0.0.1

Guardix CLI 0.0.1

Cross-platform binaries for macOS, Linux, and Windows. See the
install instructions in the README on the
guardixdev/guardix repo.

Changelog

• 0f483a54a7a9af6b5a8d09ddc04d37667c77f68d Ack invalid UTF-8 Pub/Sub payloads safely
• 04fdec9d24edd15c7cee4c5322a0ed7cd76cfb12 Adapt code-tool loaders to slimmer CompileTarget
• 6952cf5c818214b1fe0b2dc0788b9b92d1d28102 Add 'agent/' from commit '3540824c1414dd26883d677a7cb51e43cb1ae011'
• c9b484741264cfa957e790fcc3ffced731c4c14a Add 110 more blockchain security audit checklist items (Multi-chain, Signature, Timelock, Token, Version Issues categories)
• b16153149b0db4f1439fa475990dc526299dfe40 Add 250 blockchain security audit checklist items from Solodit
• 0752856aacd58c253fa5fac306a23ab5fcf0a213 Add Architecture view to RepoFocusPage
• 11f5050e66b8efb80ae3c392479349f5d567baf5 Add CLI coverage for invalid runtime rejection in deep-understanding command
• adee6a354a7c1b879787b3fedc40846ed4a44187 Add CLI coverage for resolved deep answers file propagation
• a6308635c725cb71c0d91b2f6483e8ef29de8e23 Add CLI coverage for understanding validation error truncation
• 873e016c959dd0adb4dadcf35b7be12d3f04527f Add CLI guardrail tests for file target paths
• 30f09c7b1d8a65d885f4c36e93020aad1a745afd Add CLI resume test for active session overwrite safety
• c023e3b43622c3ca1356c4e29d139267482a2763 Add CLI status tests for deep validation and expected artifacts
• ae399ab9e6923a94ae8527721abb116260717d0e Add CLI support for per-stage deep runtime overrides
• d2f617501150b5a9c384621ac65c02bacad766c0 Add CLI test for missing run target path
• 473c189bdfd827e8b3285578e61896b7c161a05b Add CLI test for resume fallback without active session
• 36816e5ec7633d6b2505a0578a5390677afc7c76 Add CLI tests for runtime normalization wiring and invalid runtime rejection
• 0c47f3a2d1cb8c4fb3de1a0af57e37f315e07e14 Add Cloud Logging structured logs and API error alerts
• 20e75bd603a04e20bc38c153d7962b43ea07a47c Add Cursor deeplink action for finding details
• 41cff694a0cd86ce43e890ce51fcc8bf17911c30 Add Cursor fix button to findings
• f07c2e5c68d0606cd76c09a74ec70a255aaaee67 Add DeFiHackLabs experiment data and documentation
• fac130e76e20699b65cf072f99503aa69a884233 Add DeFiHackLabs experiment framework for agent evaluation
• 2bf5d51e568375a06acec5db797cac7b9dc7b400 Add Docker environment for Foundry-based testing
• 4d6dd8fb53bfc9baf9c7a701d3e1aee2cf183721 Add EVMBench checklist model experiment harness
• 4a04216abbc21ba4f77fe6fad18fca723d94018f Add FindingStatus, PreValidationResult to schema
• b031eae524dbbe8664cc9c58fbca230ac93dc1d5 Add Foundry runtime for Docker-based agent execution
• c9a8ddbbbcbee4e9d77a7b33848eb5a649f5783d Add GitHub App installation flow and repository connection improvements
• c5c279bce2aa82ec7d941f0a40a06d862b6b6936 Add Guardix favicon assets
• 5f9c7c7dde4ae66ba23f82e6ed764cb6a316bc52 Add HTML report test for deep capsule fallback
• 1d047ef38a809c9116b6d4574edcb2897720872d Add JSON snapshot templates to deep stage notes
• 8374528207d99d1d2f9db761b01679394c4abc04 Add LLMClusterJudge: per-cluster Anthropic judge for sub-cluster partition
• e7bb274bb5eb5fcb18a2cd3fc18fe1dab9380f9a Add NextGen adversarial run evidence
• 7149933c293ecc4099efc871a0082261e65aaf6e Add OpenRouter key spend guard
• cd89b862990ea50f96025f309f95b079728d88ad Add PostHog backend exception tracking and alert docs
• 639ec318058ed327402480fad28ea6b73999001e Add PostHog feature flags, deployment panel UI, and deslop cleanup
• b8a5bae72cc2f07299b559dfd95a0830a008d508 Add PostHog stage metrics and dashboard doc
• a45bfc3f96db3ef0617c13cc7495a3abc172c61f Add Renft checklist OpenRouter experiment results
• be71ff2d06ab542463afb6ffba58a030fe37df49 Add Slither-backed guardix code functions tool
• c4d0b1d922eac9523adbb8eb258a3c1617b79474 Add Telegram-backed support chat integration (#227)
• e797afc26fd5dc27a93908177b9e6adf79bd9d6c Add Terraform import blocks for existing GitHub App secrets
• 9d840d4291d106fe28e0c604025c418e2cd4b2ce Add URL object URL mock for tests
• c913c0c0022d0cad085d0304c723da463dcaaab5 Add UTF-8 encoding to defihacklabs text file I/O
• 061a1248545973af4e50396375330db44d01efda Add admin audit access and first-user role
• a20510f606c2671904fa81579dc48e3d9a43bdf6 Add adversarial L1 L2 artifact schemas
• 68a6f9cdd881962878eb9ac69f175c72f463a205 Add adversarial analyzer design refinements and critique
• 3a9286c0b7e11a18dfd0fb7ce26b002729d79616 Add adversarial artifact agent CLI
• 9db8a97c4579c157b2da522128cf4dfae4ff5b82 Add adversarial artifact agent harness
• 7073da5c61748a0f3e8be43bc80ca6864fe9121a Add adversarial evidence runner
• fc7376d0932fde7962d4ec9d42c9484500fdbf44 Add adversarial investigator harness
• 3234bfdd119d0dd2c58e2ba5294b7e2139502411 Add adversarial promotion rules
• 474579507bc81d470e87a05dc7342c84d1fc112c Add adversarial proof agent harness
• ab2f484c2a1fa85f6099e7c5c721e6e225ba4cbd Add adversarial subagent schemas
• 9c0218871f6b1ebeb673612ea86f97a423e7803a Add agent registry and split shared agent utilities
• fb2c019ec2c600840a7c98831f4d43d042138690 Add agent worker entrypoint for job runs
• 0f96816218418f32fb232b9978d3f8f730fdd119 Add agent workflows and monorepo docs
• ac83f0e310150633b6216e4ad6ce320cf14b38bf Add analysis artifacts backend + security insights frontend
• 597bfed8e370a40eeea77e949e852b97c909f81b Add architecture checks and import-linter
• 7e22ad206a6f9b8779efd7af13deec179bd30565 Add architecture overview to reports
• 46c12dee4b2090f8103a303e58dff51d2cfe1c88 Add attack step seed vocabulary
• 54606d3873a9618cd364e38dc20cac13256c965f Add audit soft delete support
• 9c3880744d1519631ea6735a073917ed407048f4 Add audit usage metrics and PostHog reporting
• 5fa2e3c0455a56e9fdadd8d44b1d99a04c88c23e Add auth providers endpoint.
• ea6b991789b0c1fd1345952150737b0b60549fe3 Add backend coverage tests for gateway infrastructure
• a07912312ecb975ef13b5298161681e7049c9385 Add backend testing subagents and update workflow rule
• 72f90692d0c9a2e1b2efa08971c5f7c88aa21e5e Add balanced PostHog observability mode for agent telemetry
• c695252a11accd12795f5c49857d30e58003fd9d Add batched validation with grok/codex
• 9da45c784376806793d7e942f019c1827f3feeef Add cancel flow for stuck audits
• 96ddc847a982a4747a9dbb80a76f5e5c5f199ec6 Add canonical phrasing helpers for bounded-FV outcomes
• 8e6e9478e13a8de02a2c5ba82c4e51b772ec7ce3 Add checklist OpenRouter resume runbook
• febf23b7e20408fb566b6cea815f887f87f192b9 Add checklist baseline reference table
• da1155f7e903132e9d47edbf0b022f9402e61650 Add checklist experiment status command
• d045fc58d7a0c531613dede6bfca00fa978a1691 Add checklist pre-validation metadata
• 30b948af378c443e3f367a7e8735d8ff62eed3c6 Add checklist tool progress watchdog
• 097d6b3df46f7dd959587a3ad3165a0b8a48dac0 Add cluster-judge pre-clusterer for cross-source dedup
• 508b1dc42dfa80a97e3f9b1af076d9bd45bcfc8f Add compiler artifact and storage layout code tools
• 57f8292bc23899d207bb4a67281d507425df0469 Add concurrency control to backend deploy workflow
• fdb56a88100a6d38a5358647afb3443731396d3b Add concurrency to frontend deploy workflow
• 1b65ccc56bc5f1423190bdf1d724cfa8811fae2b Add continual learning memory state files
• c9f6544d8df9462ea1d203fb078f5ddbc494b471 Add coverage for JSON list deep clarification answers
• 9fa9f81cbefb4f3d1c50aa4a2930e352fe06849d Add coverage for deep error state reset on skip and success
• 327120d07fc1433479a22bf518bd2a7f404a15d6 Add coverage for incomplete deep understanding status
• 2c036eff2216fa7035cf1c0040154e01f547e634 Add cursor API key secret for agent job
• 3c30d770082decb5985894a31c28e07fe134dd2a Add dedicated deep-understanding CLI command and status diagnostics
• 8413785169be2d2933539b39749e9293be762c3c Add deep understanding multi-stage audit pipeline
• 9b02e415933089c6e5973b3a82cafacaea18d1bd Add deep understanding report fallback from capsule
• 3f302972b0efcb37d024c27ed5cafc5528012069 Add demo script for personal-free-audit flag
• f29aafe87424e697e4179f19a3e671a4768d58a5 Add deployment agent prompt template
• 73462f9ad024522fad8a8ed3e43c7016c6316583 Add deployment gating and balance-verified exploit loop
• 2190a1b7ea90120a3c45a32cab57b1d3db103868 Add deployment readiness check utility
• 26c26cfaace61674b535322a56a97486cee2c1e2 Add design review and liveness audit coverage
• 4205a04bdbae519b03286bd87ce8714a92915398 Add documentation subagent template
• 1a7a6b7abd6b23e62f6fa93082bf2594e5da655e Add edge-case coverage for baseline validation and missing understanding dir
• f912f6c5fe8055209f520e86f4dd6314535806ed Add error handling for GitHub App gateway requests
• 67a83b5d493bd6d6c9381bc4dd55fded564c01b5 Add explicit *.pyc pattern to .gitignore
• ac08e144c2e9cc189c44cb47a4357bb9b54ea558 Add explicit UTF-8 encoding for static-analysis writes
• b2597fd6b94210d49e45bbfca99eb715740b009f Add explicit hydration for marketing React pages
• 7fcaf9ff1cd5ddece98e7323a0e1d4dba54d908d Add exploit capital metrics and task code details
• b30538cc5bbadf3deb20b6a1fce13c55cfbbf59c Add exploit integration tests and event safeguards
• d09f77d5a24b1d326d15fd3ee76c163821e6838d Add exploit mode with parallel best-of-N execution
• b086c52108ac8f8658fff6a4ce5ba784d4ae3f98 Add exploit pipeline blog post with SVG diagrams
• 00a5ae3350f2041ef756972c89ef45075290c6d3 Add exploit run backend APIs and lifecycle
• d9c90b7127a3f034e9514c31c06f5166caa426e6 Add exploits workspace to repo focus UI
• 15884f874c3b44b21240b3d96eea751ab501c128 Add false positive sections to reports - include dismissed findings with collapsible details - add HTML styling and markdown summaries - fix exploit scenario numbering cleanup
• f117b6840fb618073fcdcbfebbbda393836e055d Add finding review demo page
• 196e0e3f36c18a76b1a8cfec0d33884710bffe68 Add finding review workflow
• f49fabca02710a71979862dea01cabea8020c288 Add fork preflight checks to fail fast on RPC issues
• 5aa5e4957d310a478d18efabf666e08eddc6204f Add fork-only prompts for on-chain analysis
• 9d47e57872a54578796daabc80eeea8abb9e6ac4 Add free audit badge to app header for eligible users
• bc5d28c498819bc4444a3a19020f8ee9d2e137e8 Add grouped audit history and tighten audit UI
• ea62b47d65b114ad8fe3e1596e3080300d0f26bd Add guarded direct checklist category loop
• 9a026ed42256a8f415a9f9af80b40ad6d87fb51a Add guardix code analysis tools
• 2ddb42db08217592c3a8d64478374fe79d298b78 Add hero social links
• a24c259d5d32df7943d13d6a22a65a7149ff06d3 Add infrastructure and configuration security audit report
• 56914c740829db38267c29dab5b37474e15c3b1b Add initial ToB security review stage.
• eaf87e780114aa7fe60465394a3148bc370248df Add integration scrutineer CLI
• fd17de9182448cd8afb955d3b458911039496105 Add integration scrutineer contract
• d929d20dc5d3c3dc7cf1f6bb4f4ae3d072a415e2 Add integration scrutineer harness
• ae50d147007fc18db6616280b595218e42fd7db6 Add lazy client initialization coverage for worker tests
• 0c1c4930e93ad66e11c146e81064c36291726bc1 Add markdown intermediate notes to deep understanding pipeline
• b0952573c97ff9fc30d98f7950fc690054feb93e Add merge metadata for deduplicated findings
• ce5f5618004d833e681e7f4fb691973cb0d7c687 Add mock payment provider for local audits
• d1e115589696d868f17e5af9fe089643474cf6fb Add more tests
• be41cbc2d719c4b196ad812308eab459850a99bf Add network configuration for EVM chains
• 8f5426c276fcd5d0585b55be136642d82c559b56 Add numeric properties CLI
• 534cf1ae8fe5114e6e0859ee46e2b1c6fbab91ad Add numeric properties agent contract
• d2c39a840c29a8cf307f6900fa2b98f37d3fb2eb Add numeric properties agent harness
• 8933fdcd03b675a55144fd620a20e0d4723c773d Add numeric properties replay scorer
• 8673f54bf1fd892ea94461d3c5df3674a105b75a Add partial OpenRouter checklist experiment results
• c165a46c297b804b579f851cdcfd771a14d9b383 Add partial OpenRouter checklist results
• 15f49001248bdd9eab66a4d0dc84b371ce81f6cf Add payment flow and domain events
• fcdef3c943166464051ea679f2c37dbca1e7f637 Add payment_status to webhook API test
• 316531c4e4878c010a4abf6cb70f1834a7885c56 Add polling behavior regression tests for audit hooks
• e51c2bdbf7ca57499956a9aafd26a6477e21bd69 Add pricing, repos, and audit wizard pages
• a773ebbee2c2053fad5da713aa6d9be8c56f8f01 Add public fixture preparation for checklist experiments
• e83080215ca9f11879c4f2593cf601572f3b4018 Add pubsub and local worker modes
• 8f8c30bcccc549e89880f678eeeea82d3febea73 Add rate limiting for upstream fork RPC in Anvil entrypoint
• 2497c2ece4677849769bf5964b6de83d4b791a46 Add real Halmos and Medusa smoke checks to code doctor
• 2145d7012226b099fe24cb8a6ae306f4cf684ad1 Add repository backend support
• 9849255e750102f7f7398f59e641e04e290640f6 Add repository frontend experience
• 26ecfe018a1f1296ca16c474f0c1371e65cf23e4 Add safeguards for OpenRouter checklist experiments
• e81e64f03ddbd8455c6ff20204d2bcf973c3f67a Add source code heatmap tab
• 8dc90ec414e1df1a7b3179a55dd86e6f0f0a378a Add strategy-based report rendering pipeline
• a70e055193d68653e6d4681d4178f88ecd28a206 Add structured finding fields to reports
• 195ec6aa794a8f6b53998519db452b397d111213 Add test coverage for default deep status artifact expectations
• 499f10d6cf36c684b32326f5ac70fa67d9f8b79e Add tests for deep stage runtime override parsing
• 8a1c77b93ebd4e1a1ddd5cc00a2c51ab61892a6b Add tests for redesigned pages to restore coverage above 80%
• aee7172214714916afeff2ae14eac50553104597 Add user-event for tests
• 99c8c72806cd84fec118b0f068e361710883d6e2 Add worker unit tests.
• fc987de885809f1a3dee596c9efa0809a7324d5f Address Codex runtime review feedback
• 1f0a9644edafbff5d05fbc9aca1732059bc5dea0 Address checklist agent review hardening
• c162e62c22dd15eff0521e68cfa04f69122dc61f Adjust donut radius to avoid clipping
• a3b02de23c444dc7c27f77ad437f4feb28d37ed1 Agent: test stage_subprogress audit percent weighting
• d0af31ed8b4c82c6daa22656f4698337aa35a45a Aggregate deep stage runtime fallback errors
• de09e9bebdcf789ffafa73b643d2d1351854dbd4 Align OpenRouter default assertions
• a0e4627c020b1f2a9fbbb92d921d135243188612 Align adversarial test with proof confidence gating
• a0736221b19b46fe2baf4af928f0975b631bd3fc Align agent timeout headroom with Cloud Run
• ac9603938befd83a78ec03a33582b426ad73781a Align audit timeout to 5 hours
• 971a56812b7e9c476c695c832943f8744e334202 Align audit total count with query filters
• 3ffb314265a47303b11fffa957aec6add6bd0d07 Align checklist experiment category timeouts
• f57676195e0e105b70bb13d9f9551bb0e900ac46 Align findings count with manifest source
• 343d2feeba0b7496d6a11ff300a96cf1272aca84 Align metrics sanitizers to reject DEL control characters
• 701edf6a0492d7cff467b116cc789eec84fb1189 Align pricing and checkout UI with new style
• 64a938221f580233d54cb925e961bab8f3e0705e Align repository connect/list UI with dark style
• 1867052b63270934c49f067d1a7a0422eb018fd7 Allow Pub/Sub topic path fallback without google libs
• aaf32dd45a247c461b0238c3ee82d6ccca22c05e Allow cancelling audits at any time, not just after 24h
• 947980456ac04a93b66b5863e287e63765a17756 Allow case-insensitive deep stage names in runtime overrides
• 6bfeda529a027228f2894da85528d38bac56dea2 Allow deep answers ingestion even when artifacts already exist
• 689c081b2b268d8f851a15240cfbe91f919c65fc Allow deep stages to proceed with notes-only outputs
• e3a42deac3081f35f1322f24b480fb8795971c5b Allow findings load on audit completion retries
• 7973cf26c9949cd582f4020325e7e00050a86c9f Allow longer Kimi checklist reasoning turns
• e5458f7003b4cb3a01b6518e3a6ff4b51e5b28ae Allow run command to use ANTHROPIC_API_KEY from environment
• cd947cba25ad60f050cac59222d3ee88ffa09e2d Allow runtime-scoped checklist reports
• 16dec5f8dc6f5e4b3be71207d0a8cdcc5d2cc6b2 Apply runtime override rerun intent through deep coordinator force flag
• c10e4700b79b74fee474c1bc275d06ebecfe6ee8 Apply stakeholder clarification answers to deep artifacts
• 7c4ab6168f29aef7b1f47c9031123211f2cfd961 Archive failed Kimi AccessControl run
• 5173f6dda09992fadf6e75d79fc05a97feb0f918 Assert deep completeness check rejects schema-invalid artifacts
• da26502830a369bca8be2d70d0ecab8fbeff5880 Auto-repair deep artifacts during status evaluation
• 4f8c17e35402ef49da70ede9691be72f6cbd4d37 Avoid double counting Claude tool usage
• 7b354561c79fa9e4971d4cdf0e2f631e20c4ac50 Avoid duplicate adversarial tool warning
• 1a833d5c040dfcb23c4d335fc458b4db385cce66 Avoid duplicate extra ABI functions
• ae896c1bbcc8b31075c4ac35f49f132a4cabd5d7 Avoid retrying timed out Claude invocations
• 364bd91c45b659563bb50558b4201c047b5c3441 Avoid running architecture tests twice in backend CI
• 50263e62cffd0def0c34e4de818da4253c5c0688 Block audit restart when finished_at is corrupted
• c2ed5d923152f25fa329ec57b88fb41a1e4bd752 Bound OpenRouter checklist category loops
• dcbc8fab3bbe8e9b7a1032082ad08339c2aeb2d7 Bound OpenRouter checklist finalizers
• 1775740bca4194501f566023a883682e755bf744 Bound POC subagent runtime
• 8ccd3771e0aecd3a6e5f6da26279cedac8ff2653 Bound checklist agent exploration before output
• 0b3905a883624e30707df2ba43402fe0ab5c1dac Bound checklist runtime streams
• 0e5a05a1b481788ce3b4c8276983f96f9a5cc5ca Bound direct checklist category runs
• e19a9cb9a4d2bd12621a988de143b53e583d4d38 Bound integration scrutineer search loop
• f94381d42339ce2b73f5e85aba163475bcdc4f02 Bound numeric properties first pass
• 8ad83519bd700ee5839edfce15723d3a104d9661 Bound proof author repair loops
• cd72c1c4e12e545aecfd50d1459f84a1083d3067 Broaden compile recovery agent prompt + raise turn cap
• 7269416916279c3aff0714633a1c2e91d332d009 Bump compile + agent budgets after live evmbench validation
• 42445df57c44e7f6dc4ec5d56138d5732eb13aff Cache adversarial witness replay gating
• 2c78b88b7e3103cd734606c9cd8ffeb0e9081510 Calibrate severity rubric across prompts and validator
• c6bf88a10d230f792d47099f33d2476f21caba3c Canonicalize Solidity type aliases
• 1cf935c6a059c6fe49b24d47a4801525e3ab5d94 Cap session ID length in normalization safeguards
• 7e6485260b965d4b9f439f714afd663d1e5ec1b5 Capture agent report artifacts for results
• 7f0e8558b91be774f3d0dc852881664b87cf3bea Catch APITimeoutError in Codex invoke_structured_once
• 4af0c9920d130b8c9f84ab6180896d0b163cb8a8 Centralize baseline artifact requirements in deep validation pipeline
• d15ff6fab58ee321f6eb2cbae8625483159533c1 Centralize required deep markdown headings in schema constants
• c3be049cbb69abea995ef3e02cd6831a74336141 Check OpenRouter spend before category batches
• 52d924596ff5234105394173a2bc2040743e0161 Clamp negative token counts when estimating model costs
• 29b00d9bccc24737948d7e74eed71f67437693b8 Clamp successful exploit attempts to reported total
• e8998b1bbf4c63f8a74e3516f15d7fddf6f760da Clarify POC timeout diagnostics
• dbafec1532692833f7afda0631c5e3a56edc5d71 Clarify adversarial validation evidence
• 0101124c1c6d22d3eb9fa11623449c43e32b4d82 Clarify baseline deep prerequisites must be JSON objects
• dffbf11a5bebc2b3bb32e7d100a6e8271ff97a79 Clarify deep heading validation requires exact heading lines
• 9f2e5a49ce84f0ff8dfdf1d3824f8ac463dba0a9 Clarify deep status missing artifact reporting in README
• 244ec1afa7c09477a43954e58c2a46f604cb741d Clarify finalizer evidence truncation
• a957415c4a979a222adc0622670cfe95c1411e7e Clarify public checklist experiment selection
• 59e68954992143b90dc2fc598fb0a2c1ef958bb9 Clean up cloned repo directories
• 02ade7ad216805e06dac701ac73a6e02f270983b Clean up heatmap navigation dead code
• 20181c056838dbff7914b8c87dc729c87a0ef8fe Clean up trailing whitespace in exploit_prompt.py
• 4c9e8da73c299b5cbcc9a603cc8fbba96cd124fd Clear SqlAlchemyUoW session after exit
• 26506f463cf734bf0c1b65ccd102c9765c98c4a4 Clear stale active pointer when active session file is unreadable
• ddbd2f6ae120568b1c17873ca9168cc3f20a658f Clear stale checklist run state on restart
• 3fafa2d1bdd213d3627eb060a7fcb09aa2dc8a88 Coerce capsule deep question counts to safe integers
• 11a90945816ba7fff1471a493515ac3016953dac Coerce exploit report metrics and validator model names safely
• 1d1c3471dee0891a5af8431795e1fee72c4c87b0 Coerce malformed aggregate metric values to safe defaults
• 95dce3451081c5dd5f0327006b732b03daafc40c Coerce malformed pricing-known flags in metrics record paths
• 6c3db066770a4d753b0f50572645d5ef8fac42bc Coerce malformed token shapes in pricing cost path
• 9b82cdfa2c83dc42e19a37e9f4560dfb43d16e26 Coerce mutated pricing rates before cost calculation
• cbab4c8819517d243cc220b19fb21b4622343497 Coerce mutated tracker scalars during collector finalize
• 563a429a7f214ec5ed56d32d2031acbf8dc13a30 Compact OpenRouter checklist finalizer output
• 1ebab84c4cd0a48b4e7e2ed2f71f8ebad3c73b4a Compact checklist finalizer handoffs
• ef2d3668ec5afa81d2a38bfa36f76ac4ce3cf643 Compile deep JSON artifacts from markdown notes fallback
• bba92f8c9d0c6704f7808de82e5d86f5eb9bdf72 Continue checklist runs after missing artifacts
• 6c47c1b7f5c1cb0ba750df77fa37e4cc4d200752 Count only structured hypotheses in audit status
• 68298f0677efcb3c4aabec2286ffae56faaf2023 Count only valid turn entries in session show output
• 2f68873e210b8f2aab6e7ad2662ed08401085921 Cover DeepSeek OpenRouter checklist controls
• 550885a4047f2071f5725f2f567cadd264802be6 Cover exploit manifest read failure in deploy-complete webhook
• b3761a9f49df080a2d485b8395b0757aed9c1536 Cover invalid deep state with full artifact presence
• 9ad072ee9daba8aa053a97d75f0859f9080e2000 Decompose codex runtime responsibilities
• c7409b685112a85c32aa9daaa4c77dd6ab9e0bab Decouple GitHub App install-url from gateway/webhook readiness
• 9a1ccac63df0e5d381361fd7d899e626ee2a954f Decouple checklist finalizer from codex runtime
• 3b2735841d70979632262a1c96f8a72895859c67 Dedup + severity-calibration replay scripts
• e57e8da66bd4a13b82eadc826c7839ba92081eed Deduplicate CLI deployment ready-check implementation
• 19afb040e16681946371a67b586ef36e62ae9a52 Deduplicate CLI directory resolution helper
• ef848defc72424f583b9101dfa3c5654e28ff111 Deduplicate baseline artifact validation logic in deep understanding module
• d23d0f1892f52fccbf9db3ff5677ca0095b4d3da Deduplicate checklist signal test
• c707814f2610504d5e2e847b803605ff09badcd8 Deep understanding: agent-side overview split, system map UI, markdown rendering
• 56ea72fae13baa100f243d8873b4a274c0d31d06 Default adversarial-analyzer stages on; opt out via flag/env/repo config
• afb4bb39edd38cb9a0dc73a58117173e520b6a3a Defensively sanitize malformed aggregate call payloads
• 3cce20e4589e04fe6ed76b58ee074f5cc0f4fe1b Defer PostHog bootstrap with requestIdleCallback
• 25a3df313a3572b4abb489c35b766ed06d585d5e Delay GCS client creation for findings
• 165dbb821a01d31b2dc655cfb0f0822b6063b668 Derive deep completion directly from artifact set and validation
• 4d2c81da0ea04ca24014a28399fd9fbdf1d1b360 Describe adversarial proof tools in prompts
• 4a88bab3babaaf94c39894c6ae596ef24c96035b Deslop telemetry event construction
• 7cdfda4f045fd12b36c9f4b9c133b4490883271b Disable DeepSeek thinking mode for OpenRouter tools
• 2456939e55e9caa23c606403752ec4bbb3d25a8e Disable Kimi checklist inspection stop controls
• 1d46829db2fe0e438d26b2184a88fab062d6a8b2 Disable Kimi fallback timeout controls
• 0d722745b28aa000fb9e210eb5028a3807fa67f9 Disable OpenRouter checklist fallback controls
• 1eb47b4d50546df4549aa877eba2df55561a895f Disable OpenRouter checklist runtime timeout
• 2e038862f28dcf9291777299013cbfbeae15a717 Disable checklist experiment fallbacks by default
• 00123f8eee7b6058a1e5784045365678ceaa544b Disable metrics collector on non-boolean enabled flag
• 0d503b01b68510b8a9683657f78f68427eec2666 Docs: describe agent structured event stream
• 39b3d3e652b23fe62f2fda633276c02fc2164a38 Document Cloud Run job configuration updates
• 4ca5bc384be7305a18caeca28b30b494ad0903b6 Document DeepSeek control coverage
• 950b38c25910202bcd8b784c20f45e9501d1d978 Document EVMBench code tool runs
• 26b7121e494bd8226fd6c50e436ae0d7e251db7c Document EVMBench scoring spend guard
• 4ad891f5d042cb469406204197d40e109eae073c Document EVMBench storage layout recovery findings
• 23e12284a898f8043bcfe1fc6d0b18960b96d03a Document Kimi checklist control investigation
• 78cf87a74c8d27257bc4b2ceeb03085aa30a0e21 Document Kimi checklist reliability findings
• b4c74fe16aa9c411edc10c1213b423f929d6b5ba Document Kimi no-stop target commit
• 71825df888fe7354026a84e89f19b1ce7557e539 Document Kimi run stop point
• 889213b5c952512f7d7434b1102f4dd5c21b70f9 Document Medusa wrapper validation
• bdf06429bf5db6726b1b7130694d39085f52a417 Document Noya Kimi checklist verification
• d5b1815bd48bf700208d12bc9744c6c327bca05a Document PoolTogether adversarial validation
• a5f14dc9b82fb57c16213bb7f33c07cf1cdf7bcf Document adversarial EVMBench validation
• 1b61db45547b35fdf9eb6445ce97c8ede6aa20ce Document adversarial analyzer progress
• fb0983e73e61058faa2b6d6b0311f010e83dd761 Document agent vertical slice migration
• a57af623fbe5066a3f77308db9c5edf93f7c0ee9 Document available deep stage names in README
• b49e4cbbd0226afdf4a59c6414a5e5deacb01d9c Document baseline JSON validity requirement for deep complete status
• 2d17d07022e8f0d428d188a11f74034d9c1002c6 Document baseline understanding prerequisite for deep complete status
• 3cefa44dfa88a795f29a9f0c1d5dad20e343b2d2 Document boolean deep clarification answers are ignored
• aa10eef30d2e14e637127e5b9aaf3813788dd618 Document case-insensitive deep clarification ID matching
• 58fdc7069ab2b91368a0037293b4d848b667ce2d Document case-insensitive deep stage runtime override names
• 23225f25d2f9db233c29bce17b56f3105d8a72fb Document checklist experiment completion audit
• cc02e96e65cf346aa210ac7d930861d049e705a2 Document checklist experiment completion gaps
• b530d48424a8beef1d4b3447b0f0cf5ef17cb8a5 Document deep answers ingestion without forced rerun
• ffb7999914614458169e7d29d65aaa06b956a959 Document deep command incomplete-state diagnostics
• 11e77518b3d70feb643d9addbca7d3f45559afad Document deep command now surfaces coordinator failure errors
• 8b50ad2109322bee45a5626c9c4ec9383868d771 Document deep markdown heading order requirement for complete status
• e392e9d449d5503c832e259b80f875e11765f354 Document deep option ignore behavior when deep mode disabled
• b4d89c48a5cd39e9cadb0c91a749f0dc865eb7f3 Document deep run fails fast on invalid baseline understanding JSON
• 6e4ce9952a2839df524ae1a0c6acf03f6a10ca5c Document deep runtime override mode normalization and duplicate guard
• 6a7e2e66027e0278e2b927fb68411ad5fa0baba0 Document deep runtime overrides trigger rerun behavior
• 3d55dd5b86729a9b74dc8b86fb55567cc0c12d40 Document deep stage override mode validation against runtime config
• 93a1d6175248475358f6ceb7231d8fbbf95e45c2 Document deep status details in README
• b668624e06327a6338238a548eb8cf5a31af6609 Document deep status question count precedence
• 78d222776085828243a1b644c25a0a490933ff97 Document deep understanding artifacts and controls
• 284d15e867a5cd2f81964c761500673847d8f0d9 Document deep understanding artifacts and notes recovery
• ff6ec523e835eab41cd8ab7bf5a99c0555b81885 Document deep understanding audit flags in README
• 0178329616ac65426645d276ee624743a1cfa7c2 Document deep-understanding CLI command usage
• 431f6d5c57ee0e7663a19a9cdb100dddd642cda9 Document deep-understanding CLI mode in integration docs
• ae18ff546f17619d61da1644e3245af6b6e70a44 Document duplicate deep heading rejection in validation rules
• 15a72b512159f09df770dc72362b397eefe854f2 Document hyphen stage alias support for deep runtime overrides
• c259a278047c842e642c8483ce9f221de10f7fa1 Document implied deep rerun semantics for stage runtime overrides
• 9a8bc594bbf24d8c3cb2a0250e763ffe8d69912b Document non-scalar deep clarification answers are ignored
• 7fd589b83c5e957598ac0d20d22eb0a450c28c5a Document non-zero exit for incomplete deep-understanding runs
• 71b180a9f6137cb474ee2aa7fdaa5bae2b64bc62 Document null deep clarification answers are ignored
• ae10fa67a2a138569dc5263c36529dd468aba769 Document prepared 10-case checklist workspace
• c82c1ef2e6bbe67f6457416702ae2ac309d2ce62 Document questionId support in deep clarification JSON list format
• 0d2c8245a8aca08669671eef5430befb793e2f9a Document repository management
• 02f466baa85c14efed3a26686052c5e654236a8c Document required existing-file semantics for deep answers options
• f248b20dfa073fbf955cab39cbab090f6d64b421 Document runtime normalization and validation for deep-understanding command
• 76523a80174ca6f7f3c2394dd32766fbfb76319f Document runtime-scoped report support
• 883557ddcbc16927c88296dd14b271ba64481f8f Document star and numbered deep answer formats
• c39e157c626030a6124787eab279d2943a115b97 Document static audit runtime validation and normalization behavior
• 1a535d14db5dda3ac72e705a3450aa7357607732 Document strict validation criteria for deep completion status
• 2a08bc834984d636669b1444aa9e44a35e24f6fe Document supported deep clarification answer formats
• 748e0dcc8bf0b78c0cf8d432fe68bf3f54a9fa3e Document testing strategy for agentic changes
• f2267b57ca6fe473c9117368a48ce4e477b5a323 Document the customer-visibility opportunity for adversarial outputs
• c56ed55c8e5cf3c45cf2a99b840b3e227d974e52 Document underscore runtime alias support for deep override parsing
• 17726cfb94dfa074cd7cff3dc2351c80d3fa4523 Document understanding status now validates baseline JSON structure
• 2a674dac95c95a73cb0419dfd9ede3e1d92e84c4 Drop free-audit messaging from marketing and docs
• 66b97af0b52947626f12069a5a09b13e7e8ec1eb Drop unrelated overview bundle hunk
• edc0eed5a8a1e6c4d5d0231e6acb6953358e2efd Emit PostHog audit started event
• c1b32dd05f4b0bcea7b433274ba7b56f8a8916ee Emit structured exploit worker pubsub events
• e83b94fb7428287b0b7dfdd41b874cf426a112b0 Enable Claude web research for external interfaces
• 68fc57a5ad8c0a10ca9dd4425fcddd820c21a3df Enable GitHub App secrets injection for test environment
• 3ad091cd0d2ab7396fdeb8661440759c6927c8eb Enable streaming for OpenRouter checklist runs
• 718d290bea2516c7b5a30f5bb20dbfcf3b9498df Enforce Kimi checklist inspection soft stop
• b8d08033f0c092526f1b08fdd0f1a581e0b03a89 Enforce checklist agent output contract
• 487bddaad06f5812f0bcfdd0452a3a8d3d8667a9 Enforce file identity when loading persisted sessions
• 12d074015ce67d166b986e0a8472ca58260155c7 Enforce lefthook checks for agent commits
• 2a528fdadc80bc35408164ab488ca0feac66abe3 Enforce repository inspection tool budget
• 9d44a32a7b2e2859e6d57053832ecc1d7cfebdcc Enforce required heading order in deep understanding markdown
• 1d890a7add2fb47bb8e26ebd020eeb34cdc0eb98 Enhance parallel runner with preflight, fork reset, and logging
• 89101a3ccf5627574f28af30bb69d9b2d5121e0e Ensure audit metrics sent on failure
• 6e822d498a2d52916c3a5477923e2ca94c693685 Ensure coverage reporting steps run after test failures
• c627e5491c3349d03d3e81a5a16341921c66fd05 Ensure truncation suffix respects URL budget
• fdc4b2030cbf8b00a08253e2bee1a75de0bc77ee Escape original severity in report HTML.
• 2b55176e2c3d8ad881540dc053c4efae92ad4ec3 Expand PostHog to native full-stack observability
• c9cc045c8391609f14bb78d76fd1dabb496de53b Expand import-linter source_modules to cover all production guardix subpackages
• 54b376494e47ee166d4e03d87d31a44361e4c2f3 Exploit pipeline: PYTHONPATH fix, admin restart, optimistic UI
• beab53a80e21b08eaa1513dc92305ff1ee847b1a Expose agents from stage registry
• 3e981eac292380a85defe35407d0fc209064cdb7 Expose deep coordinator failure reasons in CLI command output
• 3e3fb612f99f6119be291802ebcdc37cff5bffff Expose deep understanding question counts in audit status
• 20400b820adf7968aceaeb8b40844eb3da864d87 Expose expected deep artifact set in default audit status
• a96bec8be30731aaa7a1bb2d408a493278842a48 Expose implied deep force behavior when stage overrides are provided
• bc5396d59b9b25e90d81f541a0e2d352d4cc5b9b Expose recoverable deep artifacts in status and CLI
• 564bf62f27547cc51729a929b1f9dc60905bc24f Extend OpenRouter finalizer timeout
• e41e4f7c0b0f728176b312fa806b6a83ca2ffc66 Extract reusable baseline understanding validation helpers
• c34afecd0a6ba24b06a39bddc51ff9003228f4e6 Extract shared relative age formatter for audit progress
• bf3a0c6c9cf9c4e19bef686772002a09f0c59b13 Extract source tarballs after evmbench GCS rsync
• 1223e60358613a5f6fa3f7e8f6ec52a83fc5d6f1 Fail closed on checklist experiment errors
• 9975dbc3551dfc757875d7d8bebfa6fca208703d Fail closed on non-string PostHog disable flag values
• c6fd2b91eac595f22bf0cb3da985f11538a0ff4d Fail closed on unsafe OpenRouter controls
• ec4f7e2c7d0d62bb422a5fe561e02a42a3add5dc Fail closed when PostHog client state is mutated
• 8501eb760086ae5466cfc2f47530237354d44f66 Fail deep coordinator early on invalid baseline understanding payloads
• 82e995915ab1acae0479a365cb00f361cecd7716 Fail deep-understanding CLI command when status remains incomplete
• 40f45376e66e3d09f656f7190dcda97d678adfe6 Fail run command before touching sessions when unauthorized
• 8c7684e53744e78ed9474c54038975db67c15e33 Fallback PostHog distinct IDs when hostname lookup fails
• 5d702fbb5081c6bd9fc02d3ee49ab84340091f3a Fallback to capsule metadata for incomplete deep question counts
• 4ead5eb817a8396221afb2d0aff7a3337d5f977b Fetch scan audit directly when not found in group list
• 135630b757241e2d9f9a792eb4ff36bae762ff8f Filter control-character labels in metrics parsers
• 02251be0dd5729db7930286e553407f19b971d49 Filter malformed findings in worker manifest and counts
• c1f8ffb9e76a6951083ff7ee5abdb3ec6eee01db Filter malformed todo entries in session turn normalization
• 09e808b832d278cafa082b2a22057d277ee4d753 Filter redundant missing-artifact errors from deep status diagnostics
• b570a7bd0d8f5405b9448a8073cfbc3e63f6a964 Finalize checklist output after stream disconnects
• 8e2f18d70a7ee03e4533a2af1a4088d0dd3ce98d Finalize checklist runtime timeouts with evidence
• c51bc3a441bed5b9d1aaaf471c3d098cf4513763 Firebase: SPA fallback for /teams routes
• 090761e5c4765acf849dbff2ae8e4ca29a257f4a Fix Alembic migrations to use psycopg2
• e0ba4103486e6388566e7846fedb9113f47e5b5f Fix AttributeError in benchmark runner and remove dead code in evals CLI
• 4ab4863463e245e9fbf57529ef9718b3aac744d6 Fix Bugbot issues: add missing gateway override in test, skip network call for malformed hashes
• 017d7f23fe1a56f71a91e003ed51bd956bf3f5a1 Fix Bugbot issues: preflight retryability, POC filter, checkpoint snapshot
• 6878ee64e43cdcf47e299d42ddbb19f755a64212 Fix CEI keyword override for reentrancy
• fbfc5e5a665387f867df98ce688a1c0072943619 Fix CI: update polling interval test expectations and restore repo-specific refetchInterval
• d8c34e1d87097aafe58739c7534720a7ca068785 Fix CLI bugs: audit selection logic, promptLine IO streams, repo list output
• 7019428c117bae4cae1e33e43634b07036defbbd Fix CLI bugs: audit selection logic, promptLine IO streams, repo list output
• 62df49c79e5d0e6867f13150f8eff9ead49bc0b3 Fix Cloud Run audit worker logging visibility.
• 10a7307a429bf7979297de14678617678b35208e Fix Cloud Run image defaults for infra-managed workloads.
• 55ab5cf92e2bf15b076c4f6fa10d8874fff47369 Fix Codex pricing and parsing helpers
• 04cf924fdae184ecb7c27eea12582a69047f83bd Fix DashboardPage test mock for audit groups
• cff14fa4d177fe7f8f538fcba66cc2d1d292e55e Fix Docker build and Cloud Run env vars
• 5085134a914c1d1722155098cac877f97f6f22d7 Fix Firebase SPA fallback and Astro peer mismatch
• 03ae0573f1571e9717781cbc0b4521460244b223 Fix GitHub App integration for org repos and stateless callbacks.
• 09f6d6853cd1fd5e008ad0efb4e4987947dd8740 Fix GitHub app installation handling
• 4d76e5b09d813c10245feca46a383e98b64464d9 Fix Grok trace auth and add regressions
• 54a3c26d2e7c8168f0041dd6b5172e7e40731ee5 Fix Halmos contract scoping in symex wrapper
• 0704d370e481ff0c6173a82646ebb0b05c4003fb Fix Medusa wrapper cwd resolution and surface PUSH0 mismatch
• 54a0fb44164eae0e5f5f4d57b30aaab77562f325 Fix Medusa wrapper flags
• ee146414013a23c90cebf50417fd8a7cf5e7f5fd Fix OAuth state handling and CORS fallback
• c957133c39d76618089dba0a632ac9931e38e5dc Fix OpenRouter checklist runner finalization
• b92d215ad6fe8df2a95de534c51443bd48cd2f9a Fix OpenRouter tool finalizer handoff
• eb997761e3962ea58b6d6dea53f37a84a0b5a9d5 Fix POC verification metadata resets
• 7185b11442a90e2710ddf236c86d83aadf99cb0b Fix POC verifier setup and config
• 0f17aff13b101d51f2859bd6900cd3b0d43354de Fix POC verifier setup side effects
• 60a89752c884aa3b348f3e71c8ffd5a8b95fae94 Fix POC-resolved findings not filtered in validate_all path and broken test indentation
• 0e9de6690060b3ca8685a59bf42c93e7b64ba928 Fix PostHog async blocking and endpoint tracking consistency
• f4ad17001783797f595ff5766e21b8d4029fc477 Fix PostHog pageview duplication and env ignore scope
• a7c376fe95fad6235d02121d2e276c51a349ff0c Fix README: remove duplicate documentation versions
• cfe9877f5252ee21da6bc18b0047cb33483c6299 Fix SVG encoding: replace Unicode with ASCII/XML entities
• 6614035ecdb831da2c6d6aec049748ed64e80a62 Fix TOCTOU race when reading current stage in async metrics
• dd06825938a14ec5103b42d55dae6dad0ccbeaba Fix TOML key quoting in verifier
• dcdc2747175709b676cd9efe6c78c7c041d8b92d Fix UoW cleanup and remove casts from tests
• b6f05e9b1c051a53edd5208fd2102f70ff381de0 Fix admin audit cancellation authorization
• de840cf9d1b3173978fdeae3f37c42751872c47b Fix adversarial artifact pipeline inputs
• 762cd4df0e99295f177d67efefda7d8203c33f07 Fix adversarial normalization stage plan
• 310b05624f69e784b7d95bb5979360bf8c98e435 Fix adversarial output recovery and confidence gating
• d83b60a652c179412a279b4899d3543d5f9d19c9 Fix adversarial witness replay promotion gates
• 036d950d0ce5791e167f75c8c519f826ec4a7d9e Fix agent crash and add audit events subscription
• 154910f67e0830f3995f55e4d80937c4e86a48c9 Fix agent failure classification and POC resume checkpoints
• 645089dac98ba9219377b66aa16a1753a8731160 Fix agent runtime timeout handling
• 251d36472c5e184f71afe52b5761f39e07a8fbc0 Fix appendix architecture heading guard and remove dead helper
• 169500c78d5cfba57ed34f30f13f38b4bc10c9ea Fix async LLM event dedup marking after finalize
• 8930a8019c14e8786c70076e281f436a8fb48ee0 Fix async shell execution to avoid event loop blocking
• a45fe0a07ce7a1fbbb5a4a08b2a8f8f12d65bebd Fix audit cancel test and route dependency wiring
• 3c932b61f0844bb110eb4ef1248c9b94d9f32cfb Fix audit completion persistence
• 1e9e41e1412c2766b728a3fe230726691559571d Fix audit diagram urls and stats memoization
• aee27c2899fad112051c5b4bbd56f1d529866f81 Fix audit dialog state and cleanup audit service
• c1ce484410decaba4915ab6fa1b25dbe20fcac40 Fix audit filtering edge cases and cleanup
• fa32bf66755cd1685acb6a1820906dd52089922c Fix audit limit filter handling
• 0f87b02b3a8fcc8b209b2bee469e07525d3e3e65 Fix audit output path and proxy detection
• bc803050a30749926284076248c4389741c0d5e5 Fix audit progress page test mocks
• a1d8e1df9e94d4e7eb658e1c1cd6910d21eec3ee Fix audit progress, ref parsing, and list performance bugs
• 7dd001a11e61fd7d57cfea5e406497983c4b4293 Fix audit retry progress and repo validation
• 1544b9f7eb7456327507cff9afcfc6a4f6edbb84 Fix audit stats and mermaid rendering
• 6470e02177abf226b23c13c8cf7f409ceab72dd7 Fix audit status JSON handling
• f1003798eed2c6b16f90033ed8b73445d53ce1f9 Fix audit status filtering and pagination counts
• 6b573035a027aa96a7d66602490bd0628f80ff26 Fix audit webhook auth and polling
• dea2b0b8036f50b0b9f2db2473b7c56a7765ed08 Fix audit webhook retry handling and repo conflicts
• 7a60d55904abf945e5d3e3ca90de2bc0966f2485 Fix auth defaults and remove duplicated OAuth helper
• 263cb6ea3bec8dc4fb9ead3f3f5cb6c81c7ecf51 Fix auth test URL and unify identity building
• 0b7afa3537e5061cf22a1fbf03767959b1a3de14 Fix branch dropdown reset in NewAuditWizard
• 128069cf305f7a061e10f7dec374757a0d7b7094 Fix branch lookup for private repos
• 7c2f01ca5b722bac7fdb70220fe8126c58c2f7fd Fix bypass detection for git commit hook
• db3c88353de728ae5a9040e33effae1c6c8db92d Fix caching and exit codes in external interface validation
• 7b60cb0d146ea0e1ef23bfea5ff0cfe7907ba969 Fix checklist experiment fixture paths
• 2bca561227aa965693b6054bc1547db87011ad0d Fix code heatmap findings toggle
• a1128a7b64d2393582ab90ccf31c260bc52d83b1 Fix codex tracing client auth override
• d476053f167416782b680cba3a567bf5ea3febb9 Fix codex_runtime read-only shell regex and ripgrep path stripping
• ab5ad6754a9e54745eda4b34760bb9279741ae5c Fix codex_runtime usage payload, shell ACL, and JSON validation edge cases
• a751be67cf0fc40bb69b789f397065d64b14530c Fix comment filter on unstripped line, redact hardcoded team ID, remove unused imports
• bbc89497f42cab9e687542d8c8f44767547359ed Fix commit hook checks
• be7aa9d92a2248c8e67fd1672ff67d3182a75e5e Fix connect repo dialog readability in dark mode
• a7f19ed7d111f7dd47845dea55e95986fd9d5ec9 Fix container cleanup on timeouts
• 68cfcd5c447fa32c2e232d14489d08fa7c42a6e2 Fix content_encoding field dropped during documentation source validation
• 405a2d9ca56d6e2ac7f68983e82703a16010b536 Fix cursor agent runtime in image
• 855ab25bad64ddfd5f311935e7ccccbd9adf905b Fix cursor prompt truncation for multi-byte characters
• 207f3415ec613faf160f3410d31aa3d705d11a31 Fix dashboard audit group limit and login test
• a653e4ba414990512ec9b6f1f93efc07825da3cc Fix dashboard stage for inactive audits
• 610cc377954275d46517ba34f9e7cb7acb58592e Fix dead code, analytics inconsistency, and redundant wrappers in ConnectRepoDialog
• 134ea1b086a7abbbc80557b710cad1fffb8b978f Fix deep-understanding CLI status lookup import and add command coverage
• a90ffd0542db0c3c5ad85aa269fbdd470a45c1e0 Fix deep-understanding ensure and stage override forcing
• 9428dc7f3d44ee16c6a39bca9a524d521ae3a94c Fix deeplink encoding consistency and tighten util export
• 4f3c17cdf3ee3516b8377c274f44bfff6f8398ec Fix deleted invoke_json call in single-finding validator and add missing grok aliases to evaluator
• 3ca526bcc55ef2955f1ad149b8c57ad65f43158e Fix deploy tags and docs paths
• 9ad10f077b8ae892023c8a944bbfa78392577a0a Fix deployment prompt to enforce production environment
• 61fbb5b7e31201a4a6f94ac602f883f717fe19ac Fix deployment workflow and infra configs
• 328eb20b9f8a1dc165948e81e371d7d734e14a82 Fix deployment workflow ordering and Terraform bootstrap
• efa13a649e02a8c4264e2cb8ec358c6e88882730 Fix detached audit access in retry webhook
• 0a15e11b615a6598d9b417d372c97e72745dc35b Fix distinct ID normalization and dedupe checklist runner
• 2d37294b38a653f2c1ea460551263493b7250d16 Fix documentation sources: schema limit, multi-file drop, and initial audit job
• 07be48c3d343f5f61e0fdfe77b6c631199b69e51 Fix duplicate keys in attack scenario steps
• 9e8a657ee280b436b20d75344c569d890066cb0e Fix duplicate markers and differentiate resume_failed_only behavior
• 6a08e0aa8c887e9da24eb49b62bddbe062fd419f Fix duplicate project grouping and running-audit repo state
• b673eadd4cd82ed3f3a91624a7acf41acc030e1e Fix duplicated short code resolution logic and incorrect review status count
• ebec13dae68de48d2924e7d4de6c49de6f32e101 Fix empty audit report stats when validation consensus is missing.
• 773cc64c8aeb0f401edc233c916857282fca91a4 Fix execution log null/zero handling
• 97d6465c05a0883e3ca311f8713c54a7476c84d9 Fix exit code for non-clone non-retryable errors in outer handler
• 5184ff9a0953f076e9675b7708bb456dac2fe864 Fix expired repo access handling and restore PostHog runtime.
• 6aba3f74777d8592237c86b6ec873edce46b61dd Fix exploit container Python module path.
• 979b02380b6f339db1e12dda82cf4ba5c8164315 Fix exploit decimals feedback and JSON mutation persistence
• 8fd15224b14e2f3f6e3bb5c3562dece41782d830 Fix exploit deploy runtime config access.
• 3e31cbcf3ce196d7c67eb7f0762012d69d6de878 Fix exploit finalization and event handling edge cases
• 989c9739d56d03cb012b7796a5bd0b54c9c82dac Fix exploit workspace start-state and null capital display
• 2754e6fa225fc76f92fd88629674395068e41461 Fix external interface extraction edge cases
• c516a7bd63ea5d8b0ab760b6d34841a539c3681e Fix external interface location and dedupe
• 1f15443b91163814fdf1fd957877012df0fa3c44 Fix external interface validator issues
• 1a7eba3c6e83a09c75cdf6c7e7ac2d1bc734fe71 Fix fallback diff handling for undeclared interfaces
• 2da5602da34d1fdaddee744819f56deccaf858a3 Fix finding detail severity badge and analysis border
• f7af1ccae749ad52fb88154e98a5ba39d7fd6241 Fix finding manifest edge cases
• 3ac525ed3294923aa837aa8343411604eb171f53 Fix first user role assignment race
• 1eab204a98fbc96350c5977ad35659495113f9f5 Fix flaky text assertion in AuditDetailPage test
• 24e3200cb679e1f33fddb9c20462bf0158fad636 Fix frontend analytics event tracking
• 256e3519cec54c66951e2779a69c54fb370e1c86 Fix frontend tests after deep understanding relabel
• dc2d15a37fab3eb08f7522cc6c30a82860ab3a77 Fix frontend tests for audit UI updates
• ba6c5e6eb608aa0920759996e8c6a2e3af438c76 Fix git clone auth for GitHub tokens
• 9e022f00de4944c8f5be519c90b1ba0a12d783e4 Fix guardix symex reach wrapper for Halmos 0.3.x
• 379b902449257fc2d1fab1f6ecdf30e6f01d8a9e Fix heatmap back navigation state
• 232353aa7de68380d28c6f0d56fc7b1180eb7edb Fix heatmap parsing for receive and fallback
• 6027adb7d5730db31bdbcbaf2f07f2ae9e101a1a Fix heatmap scroll retry and GitHub response scope
• b51c972dfb68c01364afa391b0fb03794d0c8f18 Fix hooks.json path to before-shell-execution.ts
• 5674df2f66780d6e59a3ae54757fc0b45d2fbbd7 Fix import order after rebase
• 99ab2be9bd4a621877716eac23c8bed016fbf7d4 Fix import order in repo dialog
• 130c22c258b53e8709d7f5367d4ee4e9f6a7f99f Fix import order in shared dependencies
• de9432918f94785424284d6bf6afc701a2bad5cb Fix import-linter configuration
• 06fe72765c7fb8eef3e7f340dd8bb2ed435a4cf9 Fix info severity badge border color
• b23c9dc36aa6ae956d9f9829297307009b69f80c Fix interface comparator overload matching
• bcee8882a8e0872a62c1949d345f13df7e26dcd6 Fix invalid audit commit handling in pipeline
• 13cee0e45459b8838dc8a57f51f4ee0e5d4e843b Fix lefthook bypass detection
• 4cac70de85bfee7bbcaecd4d83a9417fad97363d Fix local auth callback origin for demo login
• 98eb5d51281130ae468a9fb2201eb4cc24b5070b Fix loopback origin detection for URL-encoded auth URLs
• 08ae20d3ef3184293ce4c6d45fe602d75b20770b Fix manifest upload ordering
• 6db969d9b5add6dabe131f353cec9698c8d93af3 Fix markdown rendering and metadata loading
• 897e40b3637ad864798a9112f43bee0c6532082f Fix medusa recovery verification source
• 0f45ea13ea5f6f8ef6e7d622da69293171844d91 Fix mermaid loader retry and dedupe user menu
• 04d0f8762d374228378d6a650e6b3b89f39400b3 Fix metrics parsing and foundry tracking
• fc24fd228f932127bb9b847a5e29f3262c123639 Fix missing subagent_completed in _run_validator error path
• aa2a0b0a0249ffbd7eb48b0dbd902ddf1acc4cbe Fix non-deterministic finding order in review flow and unify output writers
• ef3d5096e151d489ba2fa35e1433062d117fd046 Fix non-deterministic map iteration in printObject and use promptLine for callback URL input
• 3f3b091d67922f361db53f1510300a35b233a031 Fix non-retryable missing output exit code
• 43d417bd816074a8b76f2a782cc8ca3088089267 Fix non-retryable preflight quota failures
• e702c552c35017ada27cd66f4c8e6a6139f1976a Fix normalizer test expectations
• 0965b8fb2ab3f30a9c21366be4e0a8e9cc6244b2 Fix notifications session cleanup and remove dead code
• c4f3568687ca604d7bc44182cf001ca828cba7b4 Fix null progress init, dashboard context usage, and status side effects
• 7e80440de87e669b50833ac0b01cb68eaad9cd76 Fix orphan deep understanding heading in markdown report
• 6ec66660d3568a0e0ca85075c8fc0beab57297d9 Fix output stream inconsistencies and remove dead code
• 6e3916312580b74d76dfe10149734d2c2ed36233 Fix output stream inconsistencies and remove dead code
• 71fa17164df3f1cedeea46ac9526052c38f6e420 Fix pagination total and shared UoW context
• 9e4f907e0c283429737878e4930ace77f6a878ef Fix payment migration chain and webhook status
• ba50548ab4b865491eb27dc3a5b8da2a6655f437 Fix payment port annotations for runtime
• e86621ba57d6fbfa9fa0c4fa834da31ce9b53ffb Fix pre-commit checks
• 0284fc6e3eb1337484f01e26210a183778d97c72 Fix pre-push hook hanging on frontend tests
• b3c0f2088dde397ab3f70fddd17ca1f3f2bcf5eb Fix progress reporter PostHog blocking and null no-op behavior
• 4f961aaffadaa6310ea92fdef612b586cab78b5d Fix proximity clustering and family resolution
• 7163ae648429dbc1793bd82c449353d669a180a1 Fix quota exhaustion retry classification
• 47da8f6532066ed384b510980d0ca376509c3996 Fix realistic wording in checklist prompt
• 1fabde94cc7bf20d9ff0e33f04a46fed732dfbe8 Fix realistic wording in liveness checklist guidance
• eb31cad3630cc963ea97922fd3dc6d06cd0a6e23 Fix receiver matching for non-I-prefixed interfaces and sanitize ContractName in file writes
• 88d0e32acb86ba4846d452d1b14994946a515932 Fix recursive glob in fixture discovery and FP label in decision table
• f917b19229695835b8b4f16f2edb73eb93d12aa2 Fix regex for exploit step prefix removal.
• 0ce8b2d6151e7fdcf6c15a7cb1e05f89afb44fb6 Fix regex for git clone flags and type mismatch in evmbench_materializer
• 013db2d68cc716c0ec9fc217e5abded5a23cf3ae Fix relative import resolution
• 61b805fda69b92f891126d8a32532b51553a005c Fix repo focus group matching and fallback fetch
• b8ee72f25846748d50f8af59da32cf1fff2e8161 Fix repo focus loading guard for conditional group query
• bd7f4895b75c887f146377afdf20daa62d70413f Fix report download fallback and report UI
• 6f8153d2ec2782ebbd86c7b1615cda0d33c1bbcc Fix report renderer test setup
• cf1573c52d8b54e2c5c5d93062e96f91e857e4f0 Fix report tests: align assertions with TP/FP verdict vocabulary
• ddcd83743fa93bd5c1eb366946d9361061913b56 Fix repository and audit retry handling
• a916be54b8054d3cf0e0bd5d206200d918b92458 Fix repository commit timing and privacy flag
• d52775784af8ef1e9173285bbb7c7e883eacacd7 Fix repository disconnect response
• ff67ce078c2c9a79b128ddf60c4faa9852f3391d Fix repository service context handling and errors
• dc80049ada14d0750ddc0aefdb196b2789fb40ba Fix repository test interaction
• 005d15545ac2cbeb6592adc62e9b503d34159344 Fix required invariant promotion for empty enforcement
• 37c2cb61f44a06ef1f64ce76427518516c5f8bc9 Fix resume fallback and enforce manifest attacker key
• bb7c4f100550731f3e28008f3892d665bfba6f92 Fix resumed validation checkpoint handling
• a1722bc2a957fd6e7709665ec3b7d5ed79e7e302 Fix retryAttempted flag on retried API failures
• 43257c244ecfbc023fa24206c57df3895f84e091 Fix risk level for low/info findings
• 32656181f3dfe05d2899fc738843ecc428b13393 Fix role config parsing in auth wiring
• 9b171399d9927fad3960e42ceb516d6c6a4d23a0 Fix runtime NameError in run command and add regression test
• 7f5b911183326835abf22d1ee4372eb039f4ec2b Fix runtime config mutation and interface regex
• c5b108cfd6ede78fb7d6f77c9807f77341be5311 Fix runtime usage parsing across CLIs
• b07825a5ae91335cef4571e5cb1db35a36292c87 Fix severity capping and validator guardrails
• e077c304f7a7a03463558f7a677d149ec90b60d7 Fix share audit return type
• f3f5bd79254d69d623c8608d7a837d2d542df2e1 Fix share findings and repo back links
• 37fe2cc7a0146ca30f0f69f2d056e6a9bcee7155 Fix share link copy without regeneration
• 4d83cc4e62571c3b4281735ac789f700306b2b26 Fix share list validation for description
• cca20b6732ef8e19da654f6df42157090809e9f1 Fix share manifest guards and token generation
• 4aaf0e022d5c51d5facee31f100885cb242a6286 Fix share token generation and biome CLI version
• bb1e1bc137aadaab44431fcad33afbe7b6584f6d Fix short code prefix to skip non-letter chars and remove unused API dead code
• c006cf0197f60c7ff4fa4fac937cf8cb9366f7ef Fix short code prefix to skip non-letter chars and remove unused API dead code
• 2ea0ac78629aa1bd7f05d303dc79babeec693a37 Fix stage CLI adversarial gating
• b204a189ce944876786ba3387f1ca0e278819ac1 Fix stale audit timeout state handling
• 38324aec8230c11a6aca2b053a050feb37d1f4a7 Fix stale cli_* import-linter source modules
• 566acf9df73ad8deed75ade0878621558b97b27d Fix static audit runtime normalization initialization and show configured runtime
• 6983884bac514083725fda944842fc04cdd917cf Fix streamed function tool usage accounting
• 8d7d18f439069db9030c7b2d7e0201337287c7e9 Fix streaming command timeout
• 9145dff1e382429daa4fd1bf59bda566f2e07c71 Fix subagent completion events and restore raw captured output
• 9e94494b4b9a2eed0ea8c1a6ac93f40a30f89506 Fix subagent test assertions
• 1352ecb633eae395abd83c799a513f2f0f14f709 Fix subprocess returncode handling and slither-detectors --contract flag in evmbench script
• 6974e086a5f83d7ac20fb6fb226e89e1784f863d Fix swapped error messages, dead code branch, and duplicate runtime alias maps
• eb87fcb969a3b3403ce11bff99af693942360a30 Fix symex status and share project root lookup
• d82baee3399b025a5205ccf01cc813d4ff943d39 Fix telemetry executor shutdown to prevent audit exit stalls (#88)
• e725327ae24bb527bbe3e82f69ac016b6932ba8f Fix test-dir filter using absolute path parts and comment/string stripping order
• 0c6c47442fa9c396e7c63814816c4f2bf507cdd4 Fix timeout precedence in agent failure mapping
• 0cbfa09eb73114cb77f84010bf28aa330980bf48 Fix tuple parsing in interface signatures
• 48da0c6a43b3c196beb979007fb3a3521970ba3f Fix tz-aware datetime and redundant resend api_key writes
• d2cb81df183e168437caf2a6c537cfbef187d1cb Fix unbound subagent_name in _run_validator exception handler
• 3540824c1414dd26883d677a7cb51e43cb1ae011 Fix validation bugs and improve PDF report rendering
• 91c55ba6d77510512265e63e13b3d5ebe6deced9 Fix webhook endpoint paths for audit events
• 55beba2898045782b5313f833ea851a87ac69abc Fix webhook payment status race
• ad15f6f5ca7d7873e2a6574520f9b6d8d3b1afa1 Fix witness replay test isolation
• 97a3a9d8911317b96a7eb7a7395c05e164ad350e Fix worker config env and cleanup
• 849a366206bfcf06177f08b2370795c01759205f Fix: Downgrade Prisma to v6 for Vercel compatibility
• a6e5f69337be9f9c4555f346e083899a6898d398 Fix: Stop overwriting secret values with placeholders
• 230e892a29c1c8a809524e3fea9804039be0f845 Fix: guard against None base_commit in submit_benchmarks
• fd2aae716c72c2524aa00098b30f96eae9962d23 Fix: strip auth credentials from env for submodule update
• 16c925ec6a9a69efe3f41e3fa64d65a4b6c427e1 Force checklist finalizer file writes
• cb5517485366a4257988177e76fab1ba7d3e5857 Force safe OpenRouter experiment controls
• 4996de91565e3fa130756b89e1352b688a56ecac Force safe OpenRouter runtime controls
• c38f7239921613cbca0495a0887a9a74f4f7fcdd Force tool calls in checklist finalizer
• 4de9c762f6258f6e744ae6bb4dc5ad6b5955486d Format backend health tests for CI
• 3e7365a5dae92f6592195742f95ee8deaa8e87df Freeze aggregate metrics state after finalize
• ba761d08b445f370798cc5e18adca465d135eff3 Fund attacker before deploying FlawVerifier in verification
• 7971eabcab2fd7d766036ea89e9eb03a6d8234bb Gate adversarial normalization outputs
• ea0890026bb4a7d19164e1be577c28126973a2e1 Gate personal free audit behind PostHog flag
• af014eb3da56bf380391e1c2a1ec971de8cc9a0a Guard PostHog capture failures in error handlers
• 916a906bfc1b13d32758eb4e5962b3a0db9de80e Guard PostHog capture properties and hostname fallbacks
• 8a6e27027233c4e355e24a068b02ffd6d5b17d46 Guard Posthog metrics failures
• 4f4e9af76b6954034b07a2a3cab1893e03a4af31 Guard analytics calls from blocking UI transitions
• 2e03261aa772308b9d2c23137b26eca8add6ea7f Guard analytics tracking in audit report actions
• b34b0ae206655c542f6f8be9d3f9458dfcd49a2b Guard blog index against empty post list
• 6e5c1c71cfe44f96e6151fc3232dd3938db0f8d2 Guard codex runtime against missing stream fields
• 50c39589285726168ef7b9fc29ce4802c4798c0d Guard collector operations against mutated enabled state
• 77ea8cc361a079cbd9f7f957f58cc71db06e6c39 Guard findings load failures and report cleanup
• f93e0adf717cb3725856ba6816bd9283369d519c Guard metrics parser entrypoints against non-dict payloads
• 1ea7ee3ecf0246f2c391f707e561a02570fd5d4d Guard parser tool-type matching against unhashable values
• e2aaccea6882eac6415982727c2c293d3c7d49d3 Guard session CLI commands against file targets
• 30de2917e64e547247a703111e82aecb1654d00c Guard usage counter increments against malformed amounts
• 42b50d177e86e3ec3ad80189d716bf5f97366afe Guide runtime tool calls to valid JSON
• 4d8a847ec02ebe766b99055a828882d4ba675be4 Halmos Test Author: PASS run-evidence on PoolTogether prizeCount
• 3ec467d37a787c48bc67c6fb64d44b966b49df51 Handle LLM exit codes with fallback
• 02d8748b0415d624409310ec48a3f90ecd2237a2 Handle PostHog network OSErrors gracefully
• 5e9772bbf8088558d449151233955192cc90ef4e Handle credit markers before timeout mapping
• e121d068ecb2818d0620de19d57e5cbe4968ca45 Handle empty stream read errors
• 05a38d771026f7125c88f60690da604ea574236e Handle invalid UTF-8 in audit status hypothesis files
• 6ab369aed6e385bfde90c876014e513bc04af768 Handle invalid UTF-8 in design review loader
• 5d9a9489da3da1a3a0ff06df5b1e9520781a45bc Handle invalid UTF-8 in session persistence files
• 2fc24baed28f47b1e4c8082060f44906627f5f54 Handle malformed session and metadata files safely
• ba8f64236d88a19fae0b1437d42fac10711d7fa8 Handle missing audit job overrides gracefully
• ddf42d6bcc4f14e75f71ef7a351321d312041516 Handle multiline interface signatures
• 76ec235468a86163cb9c85eaa9976fb28af85711 Handle mutated non-boolean finalized tracker state
• 5b56c7dcce06d21c228055140ae31ebf5218489f Handle non UTF8 files in html report
• 8b2662f9356a260b5430298540f97f5509cf20c4 Handle non-UUID finding ids in manifest
• c07647df4c12af97e34a79ea98633a242a77c272 Handle non-dict resume session payloads safely
• 0188c4b81994127b3cd89b7e11ce2845f18e877f Handle non-string model names in pricing resolution
• 899e53760dbe2c6f85149d5699973874256b0f73 Handle ownership mismatch in GitHub installation webhook
• dbc6bb488a80858739de754e3f629395c2155c28 Handle payment token resolution failures
• 0a5d2b4a5c500002c5eaea133bed9d3d9ceb1275 Handle permanent credit exhaustion failures in worker
• 500026bab6877b2ebdad4236ebc6d26068cb3b0a Handle permanent provider failures in agent worker
• 799a1658bf6ae5dee87dffec2fbb9ac7755423ac Handle repo errors in retry checkout
• f009f0479dbae51776624460bf130fb610bae299 Handle shell execution when bash is unavailable
• ad55fe624db295ef344fb9db8b2aed945bf9e822 Handle zero before SHA in frontend CI diff
• 09022a3e13a774861089cec0dba060484e328c16 Harden CLI status validation artifact decoding
• 374ce92db7ee20913d573bb0e05d4223a57328bf Harden GitHub auth and app permissions.
• c6e21ce9c95804e97f59cd74ceb02c64652c1c50 Harden HTML report readers for UTF-8 failures
• 200e93111d85a540eda20b9bb5e103d00d413395 Harden POC verification flow
• 15d516e9e300b4184ed05b898358f10a7586ba66 Harden PoC verifier reads against UTF-8 errors
• 0711819d6734e5daf30accc9c4fe138610c18422 Harden Solidity extraction for PoC auto-fixes
• aca63f75a1c9081f0d879ead396dec1e8a2afc1e Harden Vite base path for /app hosting builds
• 29fdae3994d6938fb74ae9f06ef10d9878222175 Harden aggregate export against mutated root state
• 9e16c3e063d3696695280003d580a32c2480e9ce Harden architecture overview UTF-8 loading
• 3de867b91e7434b3f2fc1de36193642645061817 Harden audit failure retry classification
• 01c566be39478dd91e4ccdd5e4f7fc5bb1e89331 Harden checklist coordinator UTF-8 artifact handling
• 200bf4afd6eba3a5a72b0a3d1c514bbea21d07be Harden checklist experiment termination controls
• 01a1f9368f69691a66c5da9f9146ec656441407b Harden code tool file loading
• aeff3647d492af00dc73ecc2d9abbc0f53f74a7e Harden config and state loader UTF-8 parsing
• a07af358ae49c76d381ca90961bc1cd38a2e874f Harden continued audit resume safety
• 1e0e308e0af3dcde3c33180ce5879cce478589e5 Harden deep artifact validation for UTF-8 read failures
• bed78550782b2ffdded10bf1f647b1b2e2366143 Harden deep capsule loader against malformed JSON roots
• 86cc2f7b5292235b5249e84251e960e2453a9dd8 Harden deep clarification artifact parsing errors
• e92dd06f7ea8336a743f653727ccc260b8543fa8 Harden deep status completeness and count fallback
• 23fee52fe225dce5628f7f7e8b240151918b998f Harden deep status count loading for invalid UTF-8
• a28d8499179f4ed6467f3d148d6e2e0682c51c56 Harden defihacklabs index loading against malformed input
• 5fd05fe83bfa54004fe3a6dca93ca7ed814ed166 Harden defihacklabs scorer against malformed inputs
• 05597dfa4d6274ee111dc56285c826e5ac223a82 Harden deployment ready-check manifest parsing
• 8301f2e06b93a9cc2f9d5578622b5b414def3703 Harden etherscan cache and response shape parsing
• 86879bf4285119d358f121e9b3f965b23999ac3b Harden etherscan cache reads for UTF-8 errors
• 73907a182ebc345dd43ec4cbb1e0a26c7f6885ae Harden etherscan response decoding and ABI parsing
• 2b519b7d6c0834dd6c94de4e1e476203d1dfc6dc Harden execution and exploit file decoding paths
• 48f6ecf78f9527347c554b05d5492c7d2b7ee103 Harden exploit verification manifest and deploy parsing
• c79513c7f3ca877add720aa0350579095c4fb322 Harden finalizer diagnostics and source packing
• fd2d522b13a601a8a85d154c33751efa80663ea3 Harden foundry runtime state file UTF-8 handling
• ef1d519995377126687f2bbad9b607cb5116ea15 Harden initial idea loading against blank and invalid files
• 1f2ba95e43da7dc8f19be9a9617224a146f41b7e Harden interface validator UTF-8 artifact parsing
• 5377417709eb0f1096b896d4f82934f79599a323 Harden local auth to localhost redirect targets
• 87926bed9ea1d3ff80dfb6a65c3a0f9081214227 Harden metrics flows against corrupted internal state
• 69848b9d787957141429cfd353bbbb9466a8c419 Harden metrics pricing overrides against invalid numeric values
• 8b3383f1c417f9dc06b18ed07b617ec696f16e4b Harden report artifact loaders against UTF-8 decode failures
• b878c43ea393784fe20e1bd515f8b73cb58610ae Harden report renderer UTF-8 writes and failures
• df6e1f065a3a325b487d7d5a560b2fa4ff088c38 Harden reporting fallback loaders for malformed validation files
• 1547b02f7971ecfb7bc6f744b1fc216aa35aa59b Harden reporting loaders against non-object JSON
• 025ab11d363e552fc8dda1d3e9e4e79e5fd28580 Harden resume mode against malformed active session data
• 303b198dbfe4b2d50c901768ecc41d5c0233d7b7 Harden session CLI rendering against malformed session data
• 19f79a64a3831f58446df153dc377eef479ee2bc Harden session listing against malformed session files
• 4216027d1f3699439c4b5ddb11158a3a40875947 Harden session message loading against malformed files
• d5c3561490dee46bfc132cb60c1334fa9e3642d4 Harden skill runner UTF-8 file handling
• 0d30d40bd6d110a7a8af3708909a31264bcd81ed Harden subagent prompt loaders for UTF-8 errors
• c334f242777e4f5bbb2df1649e7846eac5a5364a Harden tracker ingest against mutated token and cost state
• b06d1529ec30eb4d3cd284b892714ffed1efadec Harden validator vote file decoding and parsing
• b465301c9a8831b61a7883ce59db1d172ce7ef35 Harden verifier lib paths
• 681a8e33b8636e26b82493e123ed87f9bd4ea192 Harden worker manifest parsing for invalid UTF-8
• 52b0ff6e1ba25294d87961f3f93c8c4dc17e22b2 Hide stale termination signals for successful runs
• 73ab0575f47cd43f3b243f8654bdf5b58a35c9c2 Highlight Solidity source in finding detail code blocks
• f97e300d634044d1e8803665846fef33c7cec2a8 Honor checklist auto-routing in orchestrator
• b2a1c60037a86290e1017fe5e827013feb79d16e Ignore adversarial run-evidence logs
• 30bdc447529e11c54f743ba4addced456202b8e3 Ignore blank active-session idea fallback
• 46c5048e9fcdb5909027f5d0780b302ca1c85d3f Ignore boolean deep clarification answers during ingestion
• fccbb1e1822a0f571dad7b6f2e4d4b32968f5fa3 Ignore deep-specific options when deep audit mode is disabled
• ac2465d415e9e99d170daef545b1eccb545784c6 Ignore duplicate metrics start while audit is active
• 04d9e8fffe4ad3027f85047311f844210d27fa6c Ignore empty worker progress step markers
• 048a04228f3f913ce439f1ef9fd21ea6cf5af24f Ignore invalid zero-stage worker progress lines
• 09a2d4971c2c1e59ddb97e7964efed25ec3f8d43 Ignore malformed call objects in metrics collector
• 21774476ddb9cd2fc6461874a9fa81a698538faf Ignore malformed capsule payloads during clarification sync
• 29d0d3bef22d49db5ac7e2081a6eec386f9dbe73 Ignore negative reported costs in metrics extraction
• ed9e26da817cefb423836d89fbd860b2a323335d Ignore non-finite float amounts in usage counters
• d3b0f2293592b313606c5614c898b39af38a3def Ignore non-scalar deep clarification answers during ingestion
• 8742dc619e10c1edfe789390194dadf6910358d7 Ignore non-string config API keys in run auth resolution
• 9c64388a6e6ef086748d7f9d5f233be3df6f31b0 Ignore non-string pricing override env payloads
• 36351e51bd6f064a81a648cc954edf45d83a8c74 Ignore null deep answers and normalize non-string answer values
• fbd07f297b4a75ca3b3acc025c8b8e2257a8ec2a Ignore third_party and ground_truth_workspace directories
• 568b021e478fe3805f3e0cc92470044b1dc0e4ae Improve Kimi checklist completion controls
• 31bb8d9e79f575fc5f5d7c5f11b99c793c903ca0 Improve PostHog agent observability
• 1e25304a6830a5b84af269e4155951f5d6410c51 Improve agent observability with event stream
• f4ab5c45b998428345290e68c4f713416e52e6e7 Improve agent worker event publishing and artifacts.
• 64d29c28112678b6bb6e0a49ee2daa3bd7324735 Improve audit accuracy: reduce false positives and fix validation issues
• 284509f95b959e9b03fc0b210c2f16722a789533 Improve audit failure classification logging
• 308e95db0b2d8ec692397c736ea035c867a8a055 Improve audit system report rendering and hydrate understanding metadata
• dbdfae40b6c359178d4e56d888e2b61a38ba9234 Improve balanced coupling boundaries (#105)
• 16a44c64b065d20f91a7062b16180113b5dad49d Improve exploit pipeline progress and monotonic audit percent
• bdf1d8a98d7c683d2bf234fcd0bbb7b7f828f241 Improve exploit prompt with stop-on-success and clearer constraints
• fe0a5a814cdae3c44f5260871e42ac7e04146a21 Improve finding normalization and checklist coverage
• 81b5fb0a17fb9e427a680bf5f9f9dce76d9e05dd Improve normalization deduplication
• ed85d95e94b62586333f91b198211e33ea095f5f Improve preflight provider error summaries
• 0e080fd5303ad03eb24391204b964b7558efde89 Improve progress reporting with explicit stage markers
• a1fa51367146c53e296b2f4a11da134cd5863d07 Improve prompts for availability and wrapper risks
• 4b9424167e569bd367bea59004fb3c078340fc0b Improve provider failure messages
• e6d510cebc10292d2c6c9ef182656647ed21e145 Improve repo focus UX for running-first-audit state
• 65eda44bf176f9a2db179400b1f98170ab7d1960 Improve state machine field transitions
• 4a4ded620ad9e62267632ba2ff5358dc7fb9d60a Improve storage layout compile recovery
• 474f1160b0437b2d782eacb816203a5249ee876d Improve worker robustness and logging.
• 08e630e6c7218d7f630bb9116b1f635319d0b547 Include '/' in read-only shell boundary regex
• 3a803d9dbd4e12f3be8421ad257695510d61066a Include checklist data in agent image
• 0e58ffe17978f42831d394fb450ff3176a0d043c Include deep understanding metadata in transformed manifest
• 23fbac55ce5cd3b44358a3179fe031900a1507c1 Include line info in dynamic call hypothesis ID
• 67ccb3b00d60c46eb137d387bd39c371e7e94bd7 Increase backend coverage with loader and repository tests
• fec6b55e13a7d908c9ddc2f231f2ae55c84e4acc Infer capsule metadata from deep dossier markdown
• f5273bdfc6026f91e31df4e0d1c3de31ac3a805f Ingest audit findings on completion
• 7d2c0f5284db7d913ced8c51b774d62cd13f8e65 Initial commit: Security Agent Web App
• f35d66d71735278794832ba8a0831331a397193b Install Foundry in worker image
• 86575394b2bc3f54b97e1e8aee94bb69005c2042 Install PDF generation dependencies
• cabba9cb379816023a431f414239077fb5a9e766 Install agent CLIs in worker image
• 30aa1162acacd3867c8b374577212c6201acaec2 Install proof tools in agent images
• 9ba21faaa87be7b2873d2d7297bf014cf980b31b Integrate Foundry runtime into subagent system
• 39e767725c9a83f25c37f2ab9166bf5e91090f01 Invoke deep understanding auto-repair in status checks
• 82f73fad0856151abceb5b96ba3f0d5ff404ed9b Isolate PocVerifier forge runs from target project root
• 2ca2755c28e5834988dc61c45dca3be7ce89561a Isolate audit artifacts per commit
• f56720bcca3afcb51483ef04af7ec2626f98d473 Isolate checklist experiment categories
• 8543f9f7d66ef22272b092cfe2b85e92d4ba0098 Keep Kimi checklist verification on primary path
• 803af5d39edb0abc368983e247f3bbad3f6d8992 Keep OpenRouter checklist budget handoff in session
• f553e38b903a5e66e0e565e352b5756e04c37b49 Keep OpenRouter checklist on primary path
• d25112e1343f8f9575a7af831e1fdf1a82a29f77 Keep checklist experiment on primary path
• 24372a6742e949a5f773831f6a9ffac75b1f6e10 Keep running-audit flow on repo overview
• f725af6c1f26f8ab08e463e681fc9b805c2d3830 Lazily initialize cloud clients in worker utilities
• afc8e0443febba4296db58b0baff25a65d49c453 List deep artifacts in CLI status when complete
• 715158f2c9cb6e8347848043e48e3d1d28c1b1fe Load auth providers dynamically.
• f4a72abc61a50dfe98af3274b9e9ef43654ddb8d Make AUDIT_ID, REPO_URL, COMMIT_HASH required in WorkerSettings
• cea1aacda9b35b508b83ad9afacacc2b026a7c19 Make Kimi checklist finalizer evidence reviewable
• 2810bfccceb5472b4ff9ee11a33e82a0564ed5eb Make OpenRouter checklist finalization reviewable
• e29cffbff9acbfaf9e6e2f47da2ff3e27776f657 Make OpenRouter checklist runs observable
• dd2a95e4de0988f0673ac9735e2cad37051324c2 Make OpenRouter checklist runs reviewable
• 52d221dab6ae5721cc98f79d8a0fc679ce91da6b Make OpenRouter checklist stalls reviewable
• c3cfd18685f9b27e7982a76a1bc5a958e85cb40e Make biome install cross-platform
• c1a8f0a8666256dcc47532c2288fdd382d36664b Make checklist category failures reviewable
• b55674d6a74bf10665a743135bfd8bfdc555aa2a Make checklist experiment status explicit
• 7d93c5c1021ba9007ea9ae15ba5f18cab62c11e4 Make code functions read-only and transitive
• 831b957552036608474f224cae6a6d9266740fcc Make findings refresh idempotent on webhook replay
• eb007fc83f757d1b47b60657a3806c9841b4d277 Make live PostHog LLM events non-blocking
• f83ab5aa9c314f8c002e261113c50ae1352a4050 Make metrics collector finalize idempotent and ingest defensive
• b92cfd6cb61b8832774e616e8aa1adefa0540e75 Make normalize_config_string public and remove redundant wrapping of load_config
• 445ba744fe34b18fc978b6616308d7828b0d6a93 Make payment provider configurable
• 18c9bdc49ba1a83dc40b583c7037669972028cc4 Make the cluster judge always-on; drop GUARDIX_CLUSTER_JUDGE_* env flags
• 78c41614850aca4a6fa459e95730cd7f4dc9fb47 Make worker imports lazy without google-cloud deps
• 2f249b976152673520a85c92eff19774e4b716c1 Mark complete read-file coverage
• f376ff0d082b7f620525d0628493dbfc6bc4a3df Mark malformed deep artifacts recoverable from notes
• 0c828ca04f9f2c5fc0983dc8d495d751dff622d6 Marketing: full-bleed hero and CTA backgrounds, simplify hero chrome
• d72013a6ea01ef98165e2078e5684f49fc9f90f1 Match hardcoded Solidity addresses to interfaces
• f9bcba9a0ec652a3ce441fa1de58546497ca1217 Merge branch 'feat/posthog-frontend-analytics' of github.com:xeyax/realistic into pr65-consolidated
• 8fede62571e36c8212fbd4ff7fc2eff80e00ce26 Merge branch 'main' into codex/isolate-agent-changes
• a7d5f735ce8d568098d955f00b125797eb842fdf Merge branch 'main' into cursor/XEY-1028-finding-ids-and-ordering-b1d3
• cc841a37b37579b800679a41b2d133a90b62378b Merge branch 'main' into cursor/XEY-1033-final-review-step-clarity-6d71
• 422a98ca0804efb470113a7cd41f4bc7f49c1bec Merge branch 'main' into cursor/audit-and-repository-flows-739c
• 1f6d7a9a72e5aad3a505fe7342d58850a54ed188 Merge branch 'main' into cursor/guardix-starlight-docs-and-build-skill
• deba1aa1123e49e916071ceb689aac40c0c64a16 Merge branch 'main' into cursor/landing-pages-content-alignment-fd2f
• 6382f5ca8cadb6dc5d424bbe38f57c163d298b53 Merge exploit pipeline integration
• 0456b3da93ac9e5380477f0340a69e5b10c309ae Merge origin/main into codex/exploit-memory-gating-snapshot-20260330
• 5ec049fb004b4b1a15ae15abea71c3d248350fa1 Merge origin/main into codex/exploit-memory-gating-snapshot-20260330
• 879a09e2b8adc1c573543098c9d498e079674a42 Merge origin/main into codex/exploit-memory-gating-snapshot-20260330
• a077914cb1ad8db2cf158b8fbeea0280701f11e8 Merge pull request #1 from xeyax/feat/project-seutp
• cfffaf0c0cd3d87abc187360847931a414cbfda3 Merge pull request #1 from xeyax/rework-realistic
• 77695c1ce40426381398ec5bac76352063a92f8c Merge pull request #10 from xeyax/feat/results-dashboard
• ca77cc977efc22cfff55680b58d8262b0a8ee64e Merge pull request #101 from xeyax/docs/agentic-testing-guidance
• 9d1d7b160fa541d2a233e806e2b2bfa9a856d768 Merge pull request #103 from xeyax/feat/xey-863-contract-scope
• f31d43863f92f7b3e7a04f60594d1d5b002149a9 Merge pull request #106 from xeyax/codex/exploit-memory-gating-snapshot-20260330
• 28595ad2b3115a0bb596fa2b64cae10330db4983 Merge pull request #107 from xeyax/cursor/frontend-landing-page-metrics-a269
• e5b2affe47f135ce3833e3f59c941dcc4d33813b Merge pull request #108 from xeyax/cursor/hero-social-media-links-0c62
• dda8e1577b9e959fd381f274c2b97ccc9f80ba1b Merge pull request #109 from xeyax/cursor/favicon-image-setup-496b
• e419c0e4c6eb3e376523cfea6e6c8cc79a3b6495 Merge pull request #11 from xeyax/feat/share-results
• 79d5d90d6c7fca9e92291262071686ee141ea1c7 Merge pull request #110 from xeyax/cursor/landing-pages-content-alignment-fd2f
• 59f837b890b8d7f6258df1710cae3b9414b8c659 Merge pull request #111 from xeyax/cursor/exploit-pipeline-blog-post-5a4e
• e58c4d3da970146f1d1a4c03d421415bcd4950c7 Merge pull request #112 from xeyax/fix/continue-resume-safety-clean
• 6ef4f63315f967a4182023db0a85852b2fb05e0a Merge pull request #113 from xeyax/fix/continue-resume-safety
• bde96d3447331ac620c2261f1bc8be7f55df97d9 Merge pull request #114 from xeyax/codex/api-keys-redesign
• c7ca2d6d108c562465da6726eb4aa8f1064d6ca3 Merge pull request #115 from xeyax/cursor/XEY-977-contract-picker-defaults-9c74
• 2920b560e20f380e98b72762f052b61e3da99904 Merge pull request #116 from xeyax/cursor/XEY-979-project-audit-view-distinction-cd4e
• ffaf22b8c7c2eb6902739d1cd417d48e12467489 Merge pull request #117 from xeyax/cursor/XEY-978-scan-status-colors-consistency-d4b9
• a7f503997cecc6fd05534189051c1d3065647595 Merge pull request #118 from xeyax/cursor/XEY-976-branch-selection-default-a1ab
• 20c1dd9486cc0157893defb81aae42734e837ebc Merge pull request #119 from xeyax/cursor/XEY-975-contract-picker-selection-clarity-d9fe
• eb7b93541c069ccbcd03343170112ad39dfc3074 Merge pull request #12 from xeyax/feat/audit-history
• c1795f5ca4bfc7b59a4e992a5e6c166b6f601847 Merge pull request #120 from xeyax/fix/admin-continue-skip-stale-threshold
• 72ae52b3f732bdb24e000af40f1bcc8f3d5b8bc7 Merge pull request #121 from xeyax/feat/memory-adaptive-throttle
• 618b6870fe96d71d0d346f904bb7d5be59266e2f Merge pull request #122 from xeyax/cursor/XEY-1035-audit-cancel-button-access-5cf9
• fe34f1302c15159f488184c3eb8b9439c60c7196 Merge pull request #123 from xeyax/cursor/XEY-1036-audit-progress-clarity-ecab
• 66ace95f6a7cb57accfec239917ab50786e340ec Merge pull request #124 from xeyax/cursor/XEY-1033-final-review-step-clarity-6d71
• 02725eb3fa68361c67bea23afc794fe56c84d3ef Merge pull request #125 from xeyax/cursor/XEY-1028-finding-ids-and-ordering-b1d3
• 3f13ffc727b06f588561b44c1654dff286e1ed08 Merge pull request #126 from xeyax/feat/shared-contracts
• 4b787bd9bdfc2e37db20ff0b078b992810f6bea8 Merge pull request #127 from xeyax/feat/agent-runtime-profiling
• 9b0c2af5d096c680b9acaa10c55f69758a2847b9 Merge pull request #129 from xeyax/cursor/audit-and-repository-flows-739c
• b75f89deb3e93c445dfcc6555b7b2f3247cf4d59 Merge pull request #13 from xeyax/alex/xey-424-backend-implement-gcs-client-with-signed-url-generation
• 195e45e9ccdc3d633e1a2f403892fdcbe58842ed Merge pull request #130 from xeyax/fix/billing-grade-audit-costs
• e269633fad28ae09ca4026cd77eaa0b1cb434c01 Merge pull request #131 from xeyax/fix/codex-runtime-cost-tracking
• 3e1a84702e44b03786dc159e67c2e4e12d510955 Merge pull request #132 from xeyax/cursor/-bc-043fe029-036c-49f3-bad7-7dda3eff5b34-a929
• b40ef1dcfe0328b03844481bb31b7e13ea3b1c4c Merge pull request #133 from xeyax/cursor/guardix-starlight-docs-and-build-skill
• 12505f1329bf63401a984875bdd7164eee39708a Merge pull request #134 from xeyax/cursor/XEY-1053-d708
• a696d3e19466b921361481bcfc83dba688ebbfb1 Merge pull request #135 from xeyax/cursor/XEY-1054-2667
• 77fb730aa653e859f56ad8ef0af425b25888aa63 Merge pull request #136 from xeyax/cursor/container-forge-installation-711a
• 1ea5320e1aee150e4fb6b6d01cddf67e8baeb5d6 Merge pull request #137 from xeyax/cursor/dev-environment-setup-69d6
• c28d4bd76987a9a2ae1db63fad7326a216abbcf4 Merge pull request #139 from xeyax/cursor/realistic-pipeline-issues-cf9f
• e9424327f5b3f332dc880440db13a111a9c6000e Merge pull request #14 from xeyax/cursor/XEY-430-architecture-ci-configuration-a3a8
• 43b25dd413660a896db964bcd805a151693a8c72 Merge pull request #141 from xeyax/cursor/guardix-cli-interface-97d6
• 67041e0c7092cd3e9e960ad7665be26dc0c7bc34 Merge pull request #142 from xeyax/cursor/realistic-pipeline-issues-66ea
• ff0940d29e47d60206f48955766bbe49513aa776 Merge pull request #143 from xeyax/fix/release-blockers-cleanup
• 3e97b3885902468189fe79d8857235e18f80ed66 Merge pull request #144 from xeyax/fix/clone-submodules
• 9edec57f2c032a260f9a71a808e2f15e0384f970 Merge pull request #145 from xeyax/codex/fix-startup-auth-refresh
• 2c91c4de09f927fa1413b2528590d354f56efd6f Merge pull request #146 from xeyax/cursor/realistic-pipeline-issues-ef88
• 9392268ce91ad015db451ac5b1c30b1227853e61 Merge pull request #147 from xeyax/cursor/realistic-pipeline-issues-d012
• 5d77f5f9549c880f59e7dfe3ecf0c09e3c1add96 Merge pull request #148 from xeyax/feat/evmbench-benchmark
• 3b661c31bb7f3a6ff0a10b33b4a0ed3b29726431 Merge pull request #149 from xeyax/cursor/realistic-pipeline-issues-ac74
• 82be70aba8aac91713ff379fc3ce886b0fd0486b Merge pull request #15 from xeyax/cursor/XEY-420-user-roles-audit-access-e31f
• bcfb4bc7d1021da1c27cda4cea9ec1569fb5f8e6 Merge pull request #150 from xeyax/cursor/realistic-pipeline-issues-785c
• 3f9079975ff58075f21e597f445b9e4aaabe7f7d Merge pull request #151 from xeyax/cursor/realistic-pipeline-issues-fceb
• a26d6ec978887d8a85303b6db88021b178b88cb2 Merge pull request #152 from xeyax/refactor/optimize-skill-agents
• 435deb6f83f9f6d555234c7c41fa152da56b5430 Merge pull request #153 from xeyax/refactor/balanced-coupling
• 3fb4610c423c2507fff3862f746e60537771d0c0 Merge pull request #154 from xeyax/cursor/realistic-pipeline-issues-9b92
• b5d2394933aeb099c1ca399d49e56710d22f1e76 Merge pull request #155 from xeyax/fix/new-audit-docs-step-cta
• 3e3f9fdcae80870c303d180a1d901ef0f6bbc3e6 Merge pull request #156 from xeyax/refactor/simplify-finding-agents
• f355a7049cc3025da3ddf2ea825087422fcaae4e Merge pull request #157 from xeyax/feat/finding-agent-attribution
• 532765b980902369ae78c0553dcc84fa3b38d6e0 Merge pull request #158 from xeyax/codex/isolate-agent-changes
• 16e12bf9e2b7ba469994f22510ea9d3db0bd1851 Merge pull request #16 from xeyax/alex/xey-432-backoff
• 7941f9e92bc46e0dfbfefc8e256311e9d76f8a23 Merge pull request #160 from xeyax/feat/admin-all-audits-owner-column
• 70dea6affa6907e2e7f8736b1f351c8cea14977c Merge pull request #161 from xeyax/chore/dead-code-cleanup
• f06daade5aed206e62639e002d905a8a6043df88 Merge pull request #162 from xeyax/chore/changelog-unreleased-2026-04-19
• 246eb0c96e3fc93a2bc452508d6b6d2d1a5ffed1 Merge pull request #163 from xeyax/refactor/audit-slice-rebalance
• c109ea9105a7a84442090d00aba613abb5f6742a Merge pull request #164 from xeyax/docs/exploit-knowledge-remove-cli-ref
• 0a61a3f8408dbbcb1059f7f9120fc239251880d9 Merge pull request #165 from xeyax/feat/protocol-semantic-recon
• 4d275ee2e8a21cb60f8127f5e61ba6a06710ca10 Merge pull request #166 from xeyax/cursor/realistic-pipeline-issues-36a0
• e520fc7877dfc7ac8ce4a3e72e6f85b4e376c098 Merge pull request #167 from xeyax/refactor/subagent-executor-facade
• 1de359f63e7f83397aa3d07601642e10a3462a81 Merge pull request #17 from xeyax/alex/xey-427-backend-add-retry-payment-endpoint-post-auditsidcheckout
• 35e7cbfffef99ed1527f2dfc97ca5e930d4c8a64 Merge pull request #174 from xeyax/cursor/revoked-repos-filter-c072
• 841f4cd4edbddd4caee46555fdeffb53060bf0c2 Merge pull request #175 from xeyax/cursor/realistic-pipeline-issues-0eb5
• 4887cd9da0211677eba1d03d5950747c66267c5b Merge pull request #176 from xeyax/cursor/mobile-ui-improvements-3e5b
• db18f6ab833d9a40a3fe4c703fbef79e63c64798 Merge pull request #178 from xeyax/codex/agent-structure-refactor
• b9b56a0e2b9a80a0dddd0e87bff92fca1831714d Merge pull request #179 from xeyax/cursor/realistic-pipeline-issues-8bbf
• 62a41b4932f1b157038fa93a6aab50f6cb5bad22 Merge pull request #18 from xeyax/alex/xey-424-backend-implement-gcs-client-with-signed-url-generation
• 794ef70079c2cfe1cded942d6a1a084bf463b561 Merge pull request #180 from xeyax/feat/adversarial-analyzer
• 4ce5bab680c99d4743a1a04013b0ad92c5f26fab Merge pull request #181 from xeyax/codex/unify-python-cli-package
• 657d85bbf5dde3ceeccbcb078d5b26a2c774417d Merge pull request #183 from xeyax/cursor/realistic-pipeline-issues-fcb4
• 25f3a934e05e10a80ccaa495a10f70c8d7f937ed Merge pull request #185 from xeyax/cursor/realistic-pipeline-issues-b0ec
• 3b7dabdd6e7a84bc7e9754ee368091d8dfe6dc60 Merge pull request #186 from xeyax/feat/validation-eval-harness
• 3d4a11b5e3145ad4cc9ae2848b6278d17fc58ba1 Merge pull request #188 from xeyax/cursor/realistic-pipeline-issues-370a
• 9b4be27e5841e40839e18565901707167aaca695 Merge pull request #189 from xeyax/cursor/realistic-pipeline-issues-6030
• 70a413a960da9f17927d21e0835a8896954f7024 Merge pull request #19 from xeyax/cursor/XEY-425-soft-delete-endpoints-methods-7a29
• 497721d871c808fcc9746f7427d7264557f6eddb Merge pull request #190 from xeyax/cursor/realistic-pipeline-issues-6ed5
• 38050ecd09ffe63edce8bdce78f6d718ebd54000 Merge pull request #191 from xeyax/cursor/react-best-practices-analysis-c2a0
• dc2a19ebcac514de03a835f44caf67c0daa8b859 Merge pull request #192 from xeyax/cursor/realistic-pipeline-issues-abac
• b422e6f41e3a923428e97c401cfe12df1c2267d6 Merge pull request #193 from xeyax/cursor/react-best-practices-improvements-9dc6
• d1837af85609773a1639f2a67d1c3e1cdbd59895 Merge pull request #194 from xeyax/codex/remove-foundry-runtimes
• 17547c270b863a6bf2cdd74bda7dbe0c73942d1b Merge pull request #195 from xeyax/codex/codex-runtime-rework
• 9459aa2ee5331ba8931230d71df88e5c0faaf752 Merge pull request #196 from xeyax/claude/awesome-wu-bc804b
• dc1075738e8d29819e8144992ae349509ae3e46a Merge pull request #197 from xeyax/feat/code-selectors-tool
• 2bee6e904d4c4921564bde16dddf7549977fba02 Merge pull request #198 from xeyax/codex/xey-1175-normalizer-noise
• 57f19ab3fda9bbf2e192628a79dcd943ae6ebb45 Merge pull request #199 from xeyax/cursor/7f0f49e7
• 0f10e7bc87966d42f0bfbc372b3239f3dbf58c5e Merge pull request #2 from xeyax/feat/uow-testing-guidance
• 39613943debb4943f0f64e49899789117f83e75f Merge pull request #2 from xeyax/feature/static-analysis-tools
• e94df22682906c7f6131dd7c53048c271cafde48 Merge pull request #20 from xeyax/cursor/XEY-428-dedicated-frontend-pages-89af
• 9157d946e9f7ce3eabf4a96857c6ebb432757d9e Merge pull request #201 from xeyax/cursor/react-best-practices-analysis-1c8c
• 2c3f12575b707c278ddd5b93b0cfb9a5041ec0b1 Merge pull request #202 from xeyax/claude/awesome-albattani-a5b7f9
• d7c2a5851250e742e86d9260a5b886d38ee95d0d Merge pull request #203 from xeyax/codex/openrouter-kimi-deepseek-runtimes
• 8f68f96f9442413780d21b52dcd7ced420ed4ffe Merge pull request #204 from xeyax/cursor/324aa5d4
• b97d20f2e6fe12f1a7f484984c39cc7b16aef98a Merge pull request #205 from xeyax/codex/new-worktree-2026-05-03
• d67829e60c70b59d36ae05d0ffc4552a95d314ec Merge pull request #207 from xeyax/cursor/react-best-practices-improvements-341a
• e333ae107e0dd49f9b050fa208e2c1b76f24d1aa Merge pull request #209 from xeyax/cursor/react-best-practices-analysis-0981
• 7d83bcd9b5270b1a7ad192b8ed2a71afbc4a2d47 Merge pull request #21 from xeyax/agent-worker-implementation
• 3e5c1c0f82bed180c69f07f968b04da826409ea5 Merge pull request #210 from xeyax/codex/new-worktree-2026-05-05
• 3ad54c08bb01b90c88617fdfb692f055bb857e87 Merge pull request #211 from xeyax/cursor/disable-free-personal-audits-aab2
• 334d00287a73bb3db0017d3dd965a2cd9a5447ec Merge pull request #213 from xeyax/cursor/react-best-practices-analysis-95e2
• d3f20926b367fdb7207f99067dedd50ddcb833ee Merge pull request #214 from xeyax/fix/adversarial-proof-agents
• d7d6bc985dfdecb74edc4d502604e9e2719b6efb Merge pull request #216 from xeyax/cursor/react-best-practices-improvements-5dba
• 67349e620686e84d8455e07bbabf6a8244b55307 Merge pull request #217 from xeyax/cursor/realistic-pipeline-issues-3d81
• b072ee1c625b2678c210357a429ee455bde7ee3c Merge pull request #218 from xeyax/codex/checklist-agent-kimi-controls
• f0ab6555d2212864895bcb144dfebaa9fefb9a8b Merge pull request #219 from xeyax/validate/planted-bug-recall
• 9f19b012fbf85f8177590af2e504f67c9a638b3c Merge pull request #22 from xeyax/alex/xey-435-agent-worker-integration
• c8939bb7481d7813c3957fdaa7b28e0a726d1a5f Merge pull request #220 from xeyax/refactor/adversarial-default-on
• 4c74d47f4256cff0cadea46fd56f983b18a9e928 Merge pull request #221 from xeyax/fix/adversarial-promotion-gates
• 9e8b98794456d24088ac1036805c88d63f98e527 Merge pull request #222 from xeyax/feat/dedup-severity
• 0d7d6b42ee649a8d20bfb2a6b72d4b46cffce0b2 Merge pull request #223 from xeyax/docs/customer-visibility-opportunity
• e9a6ff59b38910a9076c1a24dcc79585886f999b Merge pull request #224 from xeyax/feat/code-heatmap-treemap
• 6bdf59c87ff745847c17c2d7f6791b57ddb3d43b Merge pull request #229 from xeyax/fix/cli-release-monorepo-removal
• fa73ebe03e901660146aea26b6250456c3595c6c Merge pull request #23 from xeyax/feature/markdown-export-format
• ec094d1106b19461afa5ceeadc732991d9c646d6 Merge pull request #230 from xeyax/fix/cli-release-skip-validate
• 752a691338369137f84343aea03a829ee0fe44df Merge pull request #24 from xeyax/github-app-granular-access
• bd155333dcccdc507b4f42273a2a17a5b2f07062 Merge pull request #25 from xeyax/cursor/XEY-445-posthog-agent-metrics-5204
• 4763926a016076ffdba7f44c47194f8e738070e2 Merge pull request #26 from xeyax/cursor/results-saving-mechanism-7be9
• 4c7fb25faad66ffd694f83e4f41c4910c4d23811 Merge pull request #27 from xeyax/fix/runtime-metrics-usage
• 0d739fb3665fb8a4a987246c5a7d97290c6b8283 Merge pull request #28 from xeyax/tests/core-and-ci
• 43fe8d9bb4fedd25a2a78b74b3c770e1aafd5e3c Merge pull request #29 from xeyax/tests/static-analysis
• 103d50d82d93b139ccddd98897a09e6cf1776935 Merge pull request #3 from xeyax/report-executive-summary-refresh
• a2f977a4c848209972e2f1c47b52d124f55e1fa9 Merge pull request #30 from xeyax/tests/subagent
• 3545d79aac8518e0697c5d5a4f0e4b963c3afe41 Merge pull request #31 from xeyax/mock-payment-provider
• 20534beef3227a5d00cf4057f6a409df7ad67696 Merge pull request #32 from xeyax/cursor/agent-cursor-agents
• b72e237127d7f6cca88895671272313b51fd699e Merge pull request #33 from xeyax/dashboard-audit-entry
• 5e471728d01e36a211e6653e3c055e63b17e7e19 Merge pull request #34 from xeyax/repo-branch-flow
• 70ec812f892a6719ade16869a02f2b1fb6aae70b Merge pull request #35 from xeyax/fix/audit-events-pubsub-and-agent-crash
• 6a896771849e046a3271be20f63eaa932f9b0a40 Merge pull request #36 from xeyax/feature/audit-target-selection
• d8b6247df2f08a2deb726375094183242140d4b0 Merge pull request #37 from xeyax/fix/webhook-endpoint-paths
• 73dfe869a91657899c45ab1a83329279115d7bb1 Merge pull request #38 from xeyax/fix/cursor-api-key-secret
• 29b69d8af575eac5f83f688dec4f100d02067e3d Merge pull request #39 from xeyax/fix/checklist-in-image
• 43063f85cc6ae589487a67e6927662ca1e3b2dd0 Merge pull request #4 from xeyax/feat/rework-realistic-v1
• 953db686f3f528c02d9954dc74839d864ce01dcb Merge pull request #4 from xeyax/fix/cursor-fix-button
• b31361e9519312e4601b95b2133f723fcc810ca7 Merge pull request #40 from xeyax/fix/normalization-deduplication
• fc099d309278e20f33b692453a223665f1395fdd Merge pull request #41 from xeyax/feature/external-interface-websearch
• 64bfd5d5cd687239af02b098ab207b817dae5fac Merge pull request #42 from xeyax/feat/exploit-pipeline-cloud-run
• 40a96ef7a79d59b9fbc4f80f679b27f574ced5dc Merge pull request #43 from xeyax/fix/system-tab-mermaid-metadata
• 7a163fe1109eb96c21bbe9879fb536e34b1e63e5 Merge pull request #45 from xeyax/cursor/deslop-audit-agent-improvements
• a7f1db9885b80283a170b7ad256a0506d919431b Merge pull request #46 from xeyax/feat/poc-verification-checklist
• f15d75bd2454280c0b21e4a4c6385991c82e2e20 Merge pull request #47 from xeyax/chore/deslop-skill-subagents
• 4bf593a218091896ef8f0b37cf38f33fc46cf278 Merge pull request #49 from xeyax/cursor/system-deep-understanding-8956
• fdf7a756ed3b3e8ab9fc91571efe551e3814bb93 Merge pull request #5 from xeyax/auth-user-context
• 620c42499e86a26bab535b7923c88572d5b02781 Merge pull request #50 from xeyax/cursor/agent-observability-system-b2b1
• d7d0fd00a5d35dfca583f13f3aa677bb3fe991de Merge pull request #51 from xeyax/cursor/report-generation-clarity-and-organization-218e
• c6e6ffd4f682d6ca3b9086557f1264df29b6d65b Merge pull request #52 from xeyax/cursor/solution-architecture-review-9829
• a2a4e2ebfa052f7b11f1d306a8da12ea510a2f8b Merge pull request #55 from xeyax/cursor/XEY-423-test-coverage-target-0ab1
• 2bbf80475003d19495e80ccb831768633ded59ca Merge pull request #56 from xeyax/cursor/codex-runtime-replacement-c8fe
• faf6b6552a4a80c73243d1ddd5f5b794b222aa0b Merge pull request #58 from xeyax/cursor/google-and-local-authentication-14f9
• 4fe2d04ce1b5aec603b3648c2aff5def9a4f6c72 Merge pull request #59 from xeyax/cursor/XEY-421-fix-in-cursor-deeplinks-cb1b
• 700300e2d1736720953310aec5c012d21e95d4bf Merge pull request #6 from xeyax/feat/repository-management
• baf1bda49351b1102ecba331a67f5c5a47532f13 Merge pull request #61 from xeyax/chore/claude-sdk-deslop-pr
• 22b7139afd57fc107f2aca14c9064a78482ebf22 Merge pull request #64 from xeyax/cursor/realistic-to-guardix-00e9
• 2533801b2b747aba8c664de624045b2a865f57e7 Merge pull request #65 from xeyax/feat/posthog-setup
• efd1de817bf6051d4b55c7a28e79a364df68d11a Merge pull request #68 from xeyax/cursor/exploit-agent-integration-design-709d
• aa62f99ad8c53fe16caa00e463e78c187f9fd9c1 Merge pull request #69 from xeyax/cursor/XEY-693-106b
• af9f5eece827371174615e9272281263b05cbd91 Merge pull request #7 from xeyax/feat/payment-flow
• aeb02f2dfc0445b8fdac55d3e1a0a53cbb294f60 Merge pull request #70 from xeyax/cursor/forbidden-audit-cancellation-c83f
• 79e54bd121e12284fbe1f56771d3935ca5e7729c Merge pull request #71 from xeyax/cursor/agent-credit-balance-issue-da51
• 281a543c3376e04d5116a501ab450429790218d7 Merge pull request #72 from xeyax/cursor/project-audit-status-display-8d6e
• 1801999d2a2c8acd976da282d3cd9be2951417e4 Merge pull request #73 from xeyax/cursor/frontend-state-reload-stability-c3b2
• 1fe0c7ee04a76dc2e73d9bc56ab3cad4c5e917e0 Merge pull request #74 from xeyax/cursor/critical-bug-inspection-dcfc
• 07acbb2fc46f2c5d3bf5d4d919ebf079113006cc Merge pull request #78 from xeyax/cursor/XEY-692-auth-pages-styling-a97b
• 674f5a501af55e2ed677037afeb3da5e53c25e04 Merge pull request #79 from xeyax/cursor/XEY-441-posthog-ad22
• 4d9f2d1a1263f4a9d8d619c3aeea8bac0d4fbb0a Merge pull request #8 from xeyax/feat/audit-execution
• 2fe8dcd220b1b0bfc6b5bbe669cda73a9b99774a Merge pull request #80 from xeyax/cursor/XEY-442-guardix-cloud-logging-a6ff
• ad4f09965f25a4e784530b87e81c22c68aa1ab52 Merge pull request #81 from xeyax/cursor/XEY-689-ui-8afe
• 5135cf68f79e50b377956331fd1eedd8b099ee8d Merge pull request #82 from xeyax/cursor/XEY-651-guardix-astro-migration-1b67
• 44dfa725d2464a229d53819f3c6dee8e71eedc06 Merge pull request #84 from xeyax/cursor/posthog-llm-observability-d0b1
• 70575b8cc421f39a84357754ce3aca18cc7af79f Merge pull request #86 from xeyax/cursor/backend-request-optimization-33a4
• 4b383ed0d7e3ed89c8d6f0246c63ba3c65ccb906 Merge pull request #87 from xeyax/cursor/repository-empty-state-ux-8829
• cbd317b6b0aa6034ad10ac58d15ccfa23a160157 Merge pull request #9 from xeyax/feat/gcp-deployment-infra
• de9c2486fe3d1edf907dff723d8b82e0d885e464 Merge pull request #90 from xeyax/cursor/XEY-715-frontend-ci-coverage-threshold-a7bc
• f2f8915eafefc8f7ca6752fe31dcf54e578ca41c Merge pull request #97 from xeyax/cursor/XEY-682-agent-execution-logs-ui-4318
• 8148a4759197a9dd65dd89b285ef9cd27426b8e0 Merge remote-tracking branch 'origin/codex/exploit-memory-gating-snapshot-20260330' into codex/exploit-memory-gating-snapshot-20260330
• 18b2e2f6e3e41505c7dc0072632428d1d1ec047f Merge remote-tracking branch 'origin/codex/exploit-memory-gating-snapshot-20260330' into codex/exploit-memory-gating-snapshot-20260330
• 22a2e91035a7fb54e46277c2f13f51a6d2da182a Merge remote-tracking branch 'origin/codex/xey-1175-normalizer-noise' into codex/xey-1175-normalizer-noise
• 6e6fa6caac8e177de622150f7891a8569e7258ac Merge remote-tracking branch 'origin/cursor/guardix-cli-interface-97d6' into cursor/guardix-cli-interface-97d6
• 4d2b348da2802be7a398acc61cff57b1973b1f99 Merge remote-tracking branch 'origin/feat/posthog-setup' into pr65-consolidated
• 3738db42fc2a814529eafd71882e68284c531036 Merge remote-tracking branch 'origin/main' into cursor/guardix-cli-interface-97d6
• 16cb37d0576ae6bf8e4d40081eb823d93e5b66a3 Merge remote-tracking branch 'origin/main' into pr65-consolidated
• 5fca623aa428dcf166402227b0862c2a5566318c Merging origin/main into the branch
• 3583fb69dd183fbf5902ac927482b2d05d808109 Migrate marketing site to Astro with Firebase rewrite split
• 07c6941119ed02ff816ff56b7e6411300a0a9d98 Model subagent runtime config with Pydantic
• 0ad82af9772dd767f0f6ca29e6150d27766e64a7 Move deep analysis to appendix in audit reports
• 02bdc469da7a3178a028dda3b19f2696ca4d6f1e Move design review into agent slice
• ee822f9d831de91783c31675f6cc99a11280268e Move exploit memory COPY after CLI installs for better Docker layer caching
• 17653c58882337a4fc5c92fdd166e34df3ddaa21 Move external interface validation into agent slice
• 5101c511b27a48633224af34708c42a468f645c6 Move hypothesis schemas out of static analysis
• 9c870020867efe9310ab8c36b61dcfd337457a7f Narrow code tools to reliable outputs
• 788c3b648748b7379adaaaa151cd3aa18da0039c Normalize PostHog client API keys on construction
• 0ae8df73ed44c256de79bfa8b88232ab0836c64f Normalize PostHog client host and handle value errors
• 8c1b10dd8bc1520809301b59f179bdfb8fca07e1 Normalize PostHog host values to explicit HTTPS URLs
• 5c3fc5080199f85ade458bbecc3fe1c53ab05caa Normalize SDK session IDs across session persistence paths
• 1fe0f903391a414e7ff9aeeb34aa2c4ca3815855 Normalize UsageCounter keys before incrementing
• 62935975423b3845ea5296d82ef76489354b0fa4 Normalize and validate session IDs in session manager
• d6b1b89eeeabdc6d7f092ab5b8fcb12a647b0e0f Normalize and validate static audit runtime modes in CLI
• 1a33048fb408fcb945525d056559969d195e8dad Normalize call tracker model/runtime labels before recording
• a9cca7699e8db47ce75fc11d6dd55c17d0e21811 Normalize checklist finalizer metadata
• bc4e565beed65036c8bf094e4946fada28df0fef Normalize collector audit labels before metrics initialization
• a23567802733486f72b01d2843a0cbd733908156 Normalize deep artifacts into strict final schema
• 0b14428107480d4b8c9f7a54d4be744111b39760 Normalize deep clarification question IDs for robust answer matching
• 1c2eb75a96934623293ff69a407dcc39f418b78d Normalize deep runtime override modes and reject duplicate stage entries
• 2fc16dee944bb9d8059b5933cab9e59d5989a59a Normalize finding manifest datetimes to UTC
• 86bf716ac9cb7147a4a95830ecaa1af678809836 Normalize invalid attacker private keys during verification
• aea5c462366e2e61b522b1e38d822f612272ba60 Normalize loaded session payload fields for safer resume handling
• 75de9cf6f21b503c032f9be10b96d566d8debe94 Normalize persisted turn entries when loading sessions
• 5566d052c987dca0759dbd260a32f1f2de2c05fa Normalize resume SDK session IDs before reuse
• da4c3ce53bc47560f2bf626d833418de1e25f85e Normalize run command API key sources before auth check
• ea377372e4f28741b518c129ec9d883ff219587a Normalize share severity handling
• 5eb547e6ae50e05af9af2d5837c2a8ec374d6dc7 Normalize update_session todos and metadata payload types
• 4f5fab6804132fda2436a9d5da9d163b6f53cc07 Normalize validator model names to lowercase in reports
• 2a6dd17563b01bcc3c2ddc0b6d45d4fb800312bd Observability: test event stream single-line JSON
• 262c77f7bf4899f3469a89fa550641ffc73dba23 Optimize frontend polling and query refetch behavior
• 607e699471a52a809191cd54a812b26fe0d2d932 PDF-friendly reports + Rich progress reporting
• 6b7b0cbcf00371e7291a6fc28c1b09287e96c746 Package codex runtime implementation
• 34db07e36f36fa626e3c1192e0f9924211c1aa5d Pass audit job inputs via environment
• 586f8cf82bc3a088d9f5e6e00f2af228c99a7302 Pass fork config to parallel runner and export RPC env vars
• ece81ff1b65a65086133a2496ff93eb2a8a41df5 Patch CLI status tests to mock imported audit status helper
• 2d7cf657774d545d9146332d2f34ff51f9a57e00 Persist exploit manifest for UI; polish Exploits workspace
• baece64a53184af101b9e6fa510ae02132c7299a Persist payment before creating Stripe session
• 05ac572a541eeae557140778c70e39f3c4293a0a Plan checklist runs from security surface maps
• 9f785ebeecbf232d4742e7629ba5ef72525e5d4d Planted-bug recall: PoolTogether prizeCount off-by-one detected end-to-end
• 48b089ac9400c3d692d153363160445d9579ac28 Polish exploit run/task error UI in workspace
• 80f034076139e3a071f28df359e1eee2c1654c10 Prefer longest interface name match
• 5811f04b5d6b3343263174f5189b1810987ff68f Prefer open_questions counts for deep status in complete and incomplete states
• dbb5ceaecfd286e768ea8fe58852b4c6221defbe Prepare bounded DeepSeek checklist gate
• 99e9dddee4c73cc60c3addd949da9d0a0fe32753 Preserve checklist finalizer context and lock reruns
• 54ff3b1772a689d6e9d6dfd8842ee7f027130e6b Preserve findings on cluster judge fallback
• 27d1c21d9c77053c7f6bf3be5cfc72f30a680b9a Preserve numeric prefixes in exploit steps.
• f50f1c7672ba30458491dbd62c80f11e39b469a3 Preserve resumed checklist outputs
• a6187a4512bc3c5405176ff5b98cc514f3bcdd24 Preserve session turn counts in hardened session CLI rendering
• 6845e90f2933b75ae94aa418ef326bf5eb0ee19b Preserve subdirectory targets in analysis workspace mapping
• 1c1e7e50971917afe76e0f3a77bd5ac2b07bb249 Preserve typer exits across report, audit, research commands
• c20b1ab0989a214bd942a8ffb72d6c2b40304986 Preserve typer exits in CLI command handlers
• 7b931f8b9e07d3b765f2910fddd0b7f7073e1e04 Prevent GitHub App installation ownership takeover
• 22173bf8bbac4f905fc0fd8419fdfbf8cbdde9f7 Prevent cyclic payload loops in metrics parser traversal
• 07c5f85f2f9644a27fdddf8d7473622381dcc030 Prevent duplicate ABI name matches
• 12bcf39c4aea22ef58f5f07831e86fb61a53c95d Prevent invalid share counts and token races
• 44d72056653606b7bd210f49366cb0568bb023c6 Prevent session ID collisions when creating sessions
• 79685f97188dc1684f7e9d38d6a95f2b6f9a0a3f Promotion gate: only proven-by-artifact emissions become findings
• 8e012b7247af638b5ca549ac45dda4eec39d8476 Prompt before overwrite when resume fallback has no session
• b2fd7a662e7ff9c128fa0cf8f9d2df85b85bfb3f Propagate adversarial CLI runtime failures
• a2dbce93fa3758e78b89dbcf9940d06b44e359e0 Propagate baseline validation errors into deep status
• 2569c0ab89dc52192d01bd815a4540c0b825a3d0 Raise frontend coverage above 80 and enforce coverage gates
• 0b29a461785f0bab217697e742089579581695af Real-data validation: 3 PASS verdicts on PoolTogether PRBMath math
• b27aff2a2518a6be2af77c708c4199bfffe30924 Recognize authorized code tool modifiers
• 69b933bd6462e13f0ada8e03602ada1c0260104f Reconstruct deep capsule from markdown synthesis output
• fa58edc56a73b91a77f628a5ba2720bce0aa0d54 Record DeepSeek BendDAO partial checklist run
• 0ba239c671aa648c535ceb1f5df46d061a24fb7a Record DeepSeek Munchables checklist run
• c58e6642bbc818dae29c792c755a5b684840af06 Record DeepSeek Noya checklist run
• 91fd89985ee3b40998a2269d591ac39615510519 Record DeepSeek Size checklist run
• b83e022997b59c5db9415ff5381278ea261cac2a Record DeepSeek checklist probe
• c64b92f6dd5d54712caaf8312712a6dea02139ad Record DeepSeek reNFT checklist run
• 3466546062d7370043436c1ba330845331a3bb1f Record DeepSeek soft stop verification
• b2022f53c05f4fdcdce2881192a7904634b43b5a Record Kimi AttackersMindset retry
• c999aee1fa40b60d5f19d641a3088b8ca8d36eb3 Record Kimi checklist verification
• 86f80a1b48ace78751beb308937ec169fb207248 Record Kimi full checklist failure
• 57cca62077dc7432df02d934c8dd994e5fc3bc3d Record Kimi inspection budget failure
• ca8dd100f2c94c8b69fde4f36eedbf9fb81962e8 Record Kimi no-stop checklist run
• 771df7925221e41bf39eaba5cc2183923af7446b Record Kimi stream interruption checklist run
• 5684d6bc7a9f6e1e13c387cc78f7273f8a95a6e3 Record OpenRouter checklist spend state
• 479b4c5380debd6b4ba46dfab3eccaf4d3fb3f8b Record checklist category progress for stage runs
• 06f31088fe5a45f889c22966b8486914a01eb2db Record checklist experiment status
• a544e6879520cfe48a8533c6483e7ea6adaa48c3 Record completed Kimi reNFT checklist run
• b13f292ba3d888bd15b776ce26d600f402c59e93 Record guarded Kimi direct loop failure
• ba27e58643d7ab07c8ae0c7c99390b8e836b1d72 Record latest OpenRouter usage check
• 7e617b02edbafb77b2eb1bb187093e07fb4b4496 Record stalled Kimi checklist run
• e394784761a7e300b298eb71b88d56f74f2c0c44 Record successful Kimi checklist category run
• e33fa575ad902ad8c934cbd588c6e2a522bd9269 Record successful Kimi nonstreaming category run
• 38718cfe0af2598fa39d1bddad8bfaca55856d9f Record successful Kimi soft stop verification
• 8f8ec595d682cd949cc8bfc9c1cc67f234792e13 Recover NPA output from guardix wrapper envelopes (halmos + medusa)
• e0fff5a0354a917a98ce25a1601ee180035ae1d7 Recover OpenRouter network stream drops
• f7a7d1f4ae75b5ddb406ce69759688da1e895274 Recover from malformed runtime tool calls
• 374530d30e8ec50f9bfa5c393f4eae08d8c1280d Recover numeric properties from Halmos runs
• b0789e1ae3ff494f16a5dd76293dd63ba6a919c7 Recover session updates when persisted turns are malformed
• d73f2dcc35c95fe0df6b6f6f03f472ce759469c2 Recover valid checklist output after runtime failure
• 84220f0787f2577cd881aa847501dee7142c4447 Redesign Astro blog pages and validate login routing demo
• 72140fb9d94e9caa1ba88ed49ef990d62b40e31e Redesign EIV as fully agentic stage with CLI tools (#102)
• b48ce856ba7973e85d9924d617b2964ddbc1cbf0 Redesign PDF report styling for clearer severity hierarchy
• 45056a0f1f211a6533d8f45677884b0e1a8d704c Redesign auth/select page with centered single-column layout
• 9aeb2de24f459f4a1f648cbafe8351a038517d82 Redesign executive summary overview
• c7110bebe7a69970f02ac7f9cd4038328917c751 Redesign finding detail, repo overview, and dashboard UI
• 5919ecd8bde513147740fd5ec9f2b339443739ca Redesign frontend to minimalistic Cursor-inspired UI
• 61bcd5d78088c944e6f6c772089cef54845366c8 Redesign review step for clarity and readability
• 15d024f2835da27f29591a283c231177c979232a Reduce audit polling load in running indicator
• 5bd514161e9441631a1350c72e30f5c809846282 Reduce false positives and improve finding quality
• b95ec118a132ef3ea29f8341183c93c643947f72 Refactor CLI target directory validation into shared helper
• 6bafc398ef5d10799a61127233b0ff871fa4fc77 Refactor backend testing structure and enhance workflow rule
• a8f94b6e6b1e60ef66d8b8673fc5675aa73390ea Refactor deep understanding to markdown-first pipeline
• 0596300bba330586191a902dfb63fbb1a410587c Refactor github token resolution utilities
• f00ac8fed0a82e596cb8acd69892d7e9a6a9f774 Refactor repository dialogs to shared dark UI classes
• 7dc048c6807f68123e6d2fcd3199ebf705c4136e Refactor session CLI directory checks and add missing-path tests
• 46e16178c69f090ffc1896abde19bb24fb1ff4c2 Refactor stats usage extraction
• ff4c7045ad6aa8f9b27150542ba1ade2a0226ab2 Refine audit execution log rendering
• 6258bc35b17c83c21e703d3a58aab42887bc7ffc Refine color scheme and redesign audit overview UX
• df25a43f934845ab6b11ecb539ddfd5129081fdc Refine deep artifact presence reporting in audit status
• e085d45cc581f743d6e24cfac3873714bf704546 Refine deployment prompt for clarity and consistency
• 058693626cfd77b0a1bc876e840bac713a921c49 Refine landing page messaging
• b342b836111275f645a4186fb929055e9873babc Refine repository wiring
• c73652ec616a9ad6ec3d16f6e0650a2de92b2efb Refine semantic dedup by vuln family
• 11144802113afcde55e7f643c6cbedde858d757f Refresh Kimi checklist investigation notes
• 64876583ea973857969ae490b7d079c0abd54d19 Refresh audit UI and unify brand header
• 846936e2422a4969cfde6a5cf03a3557b9c75283 Refresh checklist run state from live logs
• 5c44d93d048be0aa69daaaab1c646f0be0f48e28 Refresh the landing page layout.
• 020760ecb080e5ebe4c8cfdee24de8f15bfa40d3 Refuse checklist fallback controls
• 6b20db923ddb35479faedfa78f426e0894ec7dd5 Register adversarial L1 L2 agents
• 76eb3d21af8a5148e89baedb667ac635da281c08 Register adversarial proof agents
• 3c788364b9b5adb357b524796a6cf6765d00331a Register integration pytest marker
• d5f09d7f15abf77e4d004ea4f98a85fed54ffa69 Reject DEL control characters in PostHog sanitizers
• 987989bfa86cc94f6132f7b9158a2c11c33ee0e8 Reject PostHog API keys containing whitespace
• 1a68b272864c01a9253b6a076668fe22cb74abc4 Reject PostHog hosts containing whitespace
• 5a9cd3f235f21d1afb011d630b8ef17567061865 Reject PostHog hosts missing hostname or with bad port
• 36d8d10fbc3477be43478972f5b0ddcf8f330fe6 Reject PostHog hosts with embedded credentials
• eeed13ccf4ea8cf3fe230f057256a1332067c7b3 Reject PostHog hosts with path or query components
• 847e4a78105cc1d94e752071df2fbf2e46e64d2d Reject boolean and fractional token counts in metrics parser
• b3dd02c758605576ef8fce45526973ce3a5e9a16 Reject control characters in normalized session IDs
• 0d6f8d572cd5871ccd56a8dc900f92d3ec818532 Reject control-character PostHog API keys
• d99b564c9d364698d2d8ac6964ffde49810f69d1 Reject control-character PostHog distinct-id components
• 9587f8a056b861407aeead69b3b1c5acedf93fe9 Reject control-character PostHog host values
• 6acb9f8924975b9a92f18def654f0085a3940fa3 Reject control-character SDK identifiers in session data
• ff4c266a9d3965fb41f6ab0ad3e7d7e3134b2604 Reject control-character capture identifiers in PostHog client
• 4a7ce4cf04a16a86bc270754ba8e1b818f1dafda Reject control-character integer strings in metrics coercion
• 9b31ed52e39f555347765cb12cdbc8079ff64ad6 Reject control-character integer strings in parser coercion
• 8369b1961de984287854b2e0c47a10b2fc38419d Reject control-character labels across metrics aggregates
• 343908904d6f51981e0da1a8c1c964c4b77c4511 Reject control-character model names in pricing paths
• b2bc1cb3d4ca412b021de7e2828ee9271cbf1cb5 Reject control-character numeric strings in metrics coercers
• 70e31d4a15e87dc47bcc1ea94898d5d939a0e734 Reject control-character usage counter keys
• 9f8c7a6744ce6c37c82565a718149fb5bafd7ed4 Reject dot-segment session IDs in session normalization
• d46d4c7c0612b9a5ebc26d78742e2dfe498447e9 Reject duplicate required headings in deep understanding markdown
• 7f6e800ac406f95a8fff2266d7c47b8a762c875f Reject empty trust model artifacts
• 88ddbb256fdcbfd56e7abd16708b0e486632f1a7 Reject impossible worker progress stage counters
• 6463f05cd4053abdf0fffd1714e833c929e980ae Reject incomplete checklist finalizer outputs
• f00d68513ac8999f28b3074dd0ee15e9e9613622 Reject negative token counters in metrics parsing
• cc928d11c716b9cbce5d06234a1cc8f6e10523ba Reject non-directory targets in run command
• e27a03b474afe386730e24915aed6a5dfd9fd092 Reject non-finite metrics payloads before PostHog capture
• 9ae7a5db2744bd19a7dc769b7832c5ef69da3385 Reject non-string PostHog capture identifiers
• 901a63bc9accb3328d1f8e66912e51e4a70480c5 Reject percent-encoded netlocs in PostHog host normalization
• 930763cd4231027e36406568eebf0f86d5d52c7c Reject unbacked numeric hypotheses
• 3cb9cb14c247a0c2b5a1236ce1c64fea01516b20 Reject unsupported PostHog URL schemes and unserializable payloads
• c1cb579b7b07d06c92df5e7601ae716ba79428a5 Reject wrapped control characters across metrics normalizers
• 05a9c3c6f74eca7223cb89a8aab0daea2449fc65 Reject wrapped control characters before PostHog normalization
• 6957258d17acf1df02c1aca776140c1047f258c3 Relax OpenRouter checklist closure bounds
• fe81bd651225fb85205f091ebbde7bf362af25b2 Remove AI code slop from comments
• b23ec38d5ad8fc5ba2079d7f8de47ca46b2b8da2 Remove BugBot rules and trim hook comments
• 6f81eec95399f683fe8462d4d36e667c4d4185df Remove Foundry subagent runtime modes
• 6dbec0463668afc40e3f22957ca42ccb2ad73fa9 Remove pycache files from git tracking
• 998be07090768fc591cbc485cf9c096179eb58f1 Remove catch-all Firebase SPA rewrites
• 2c777ad3975c8dd6b436c79375437e6e618cc280 Remove codex runtime finalizer recovery
• f9b1d4a37ad66c7177495b02299ccaa6e28f644a Remove committed adversarial run-evidence logs
• 0450ca3dd92a076f98c43f71c49b919b07dbecdc Remove committed benchmark run artifacts
• db2e5114670acea1befb11c87d3fd6d93d35d062 Remove developer local path from DDD discovery metadata
• 6b7a8c9b17850f9ba27fcf4e4e4852dc3650f730 Remove duplicate _dedupe_preserve_order from evmbench_materializer, import from materializer
• d8662b1aaa02fa18116823974ccc9625b39eec25 Remove duplicate _normalize_api_key, reuse _normalize_config_string from config
• b488b6e03c9c5ae80276e06be3cb4c6079d1264f Remove node_modules from git and add to .gitignore
• cdf91a52d201851825b9630603227d0316a9e2b6 Remove redundant gitignore pattern (trailing-slash duplicate)
• 6db21940761cdacbfd416cf85c6f4429bf34b112 Remove redundant installation user_id reassignment
• 89feebba20d609631b64b3191981e02b281abb12 Remove redundant severity label transform
• e26bd81bc7f62740fa55a4540c384e694a9e1db7 Remove regex-backed guardix code tools
• f2230810d1f51667186c58671765546b42577c5e Remove temporary finding demo page
• ad1b23b6e1a23d41b0477244f5d25f2bbfb32f38 Remove unreachable else branch in _detect_proxy
• a5fd62d08bc614de6463f330782ac315e23f0a19 Remove unused env block in deploy-agent
• 44c4d00a60054348fa8272134723e1a589f0315d Remove unused heatmap contract helpers
• 386d4185de4dc384091d5bc905ab9ec90dbfe7f3 Remove unused pytest import
• 961b7a8f1e72e3bba0126e6166370868f9d83dfb Remove unused static analysis test imports
• 963e2ac4f9da87d6da95d16311840cbccf7b4b21 Remove unused test fixtures and imports
• 8a73949c7030b6b43672f04d5a09f4baa08c1a65 Remove unused test fixtures and imports
• 3b66d6ccb8aefd1e0542aae0b96059360c701a5d Remove unused test imports
• 7023a4629a4bdd62dd3c3770666ca217324bcb9b Remove unused tmp_project fixture
• 979404b128232ff78e7873c6c492f4b2ca7a207f Rename hypothesis source reference
• d833d69071bda16a891fc0ea1029706354f8acc0 Render deep lineage references in audit reports
• 15e4e2c2f0174fc52c0654ca2edd61c5419da46a Repair invalid finished-at metrics state on finalize
• 5fee6202883796d93a3fea02def48e8c7399a870 Repair malformed skill outputs and make interruptions explicit.
• cd9a8c37476171abb450428e43a44070efa47762 Repair mutated aggregate accumulators during record call
• d0c3fba00d806f45c91844e998aba0e0f407ccd6 Repair mutated tracker tool counters and collector references
• 6799a80c59bd4f9e18401114c2a0e19d2d3dffcf Repair session identity fields during update
• db030696e7f4b0b3988cfccc98dc17a895b7885c Replace codex CLI runtime with OpenAI Agents SDK
• 894a8be2fde0b16d4582979c94cd2616cc9917b9 Replace heuristic compile recovery with agent-first orchestration
• 21e713e1e39eeb6128dcbe40a8439d7e46ce9a4e Replace severity pie chart with SVG
• 4e14a3b89f2c066677e21518c6bb704fe29a01c4 Replace single ToB stage with parallel skill subagents.
• eb46cb3c8ab6f25fed01316836a09b49b89dc207 Report active checklist runs from locks
• 3362a8969c5aff0260734ac9bb7c7f2aac403d8d Report baseline understanding artifact presence details in status output
• 8d4fd55a2aaeaed96b157a63b60a1fb4659e9eab Report checklist experiment failure status
• c520e0a1f3a227575ac9a2ad9d8b5d1c957d7ad0 Report live checklist runs before stale results
• 077927389c784c1f3cbaa2a8c224f6ee54c977e5 Report missing deep artifacts even before deep run
• 86b1111720f182bdf2ed624bbe05651bb8acf13c Report open and resolved counts for incomplete deep status
• 2832ac32461726baacad492b2c963ba9890e84ae Report provider spend in checklist summaries
• 06d0084563182961872604fbbdccb322c945b3c4 Require acknowledgement for paid EVMBench scoring
• 29fa2feeb8f5f94901711cea2bf3824eb6b34b40 Require active session pointers to reference existing sessions
• b26cc7b13b6a95e3aa3de87ed4e30f963b67f88f Require baseline artifacts for deep-understanding completeness helper
• 388965cf47149213bd6617c5e61af836d3b43d64 Require baseline deep prerequisites to be JSON objects
• 92941bdfd2a9ac5ba05e7cc725db7727c0ea4f03 Require baseline understanding completeness for deep complete status
• 3c6816b1ab163a3600768736d508d58d7a39d12f Require continuation reads for checklist inspection
• 4f67bba4218cddd4aa9a8784839fba1e1d05242e Require deep understanding markdown for completeness
• 097126294b7882d5c440095596a07446867402f5 Require explicit OpenRouter checklist scope
• ff64b4b4c4378f4e3e08cbfb1cae5474051e35a9 Require explicit trust edge attack classes
• 408f7906fe471596885c1a9e2f86d404b9528921 Require trust edge attack classes in schema
• e364708569a813b9197d11db8462c3ae8bb9b840 Require valid project metadata for session project existence
• e756916de79ecd6e61a3679fcb4048032ec53ac1 Require validated deep artifacts before skip and status complete
• de49770f3dd44e084e265e1cc889c1b14d53f77f Require validation votes before fallback true-positive reconstruction
• 620b014f9ef6ff4469cfd80638f810ef3ea984cd Rerun deep stage when runtime overrides are provided
• ce55d373afda145a4eff37444d168d5168324fe1 Rescue checklist finalizer timeouts
• d30a43c83b34f22786061fb8f365a6674d125cb7 Reserve finalizer time for isolated checklist categories
• 801f748d8bf8d4aa7625db21f8d9c973c372ba43 Reset audit state on retry
• 2591ead040e4171c6968b65d23d9abd58a6a53c8 Reset audit tracking refs on audit ID changes
• c576ccb54196bdc3fecd6ce8cd6082f1e17b3bf7 Reset checklist experiment run state on rerun
• c9abcc1ad1e5c3407cd8c1c5fdf5683b9fab258c Resolve remaining PocVerifier reliability findings
• 9eccb5c57a72e8773c7c5e8afefa74ab16ccdb4e Restore Cursor beforeShellExecution hook config
• fc1eb255a808ba5305bfdcae1e13f1ce63a88a73 Restore bounded OpenRouter checklist controls
• cc67cfaf63c1a362fbb880b5f03fa02c79758ac6 Restore bounded OpenRouter checklist controls
• 82afe33b9e72a16a0cb258663303831797c4c255 Restore checklist progress cutoff
• f0f38f467cafb586d809c656870fe95afe84b235 Restore refetchInterval on repo-specific audit group query
• aae0edd0897e99053d326ea253cb5913f4ff8e32 Restore safe OpenRouter checklist controls
• a05716fe9e18a5956d84e5a2173484407cc6cb02 Restore static audit runtime normalization and add output regression
• 4bb80d8e3c92f2d841966e1fe18440511cc43608 Restore typed dependencies in routes
• 14419a370a2042a90d91084bf141656e91ae9060 Resume checklist categories from surviving outputs
• 5149b873f0a588e08ee4b295c88a8e51bba92d98 Resume failed checklist category loops
• 9ee894507ea3dc5b105e3d5ec0108d69422d8d19 Retry checklist stream interruptions
• c7316e9c4b810254121fe52a2fdd8cfcbd08ec00 Retry stale audits from resume checkpoints
• 679ee3482668f29ed22d6292c7b33c61c78aa620 Revalidate PostHog API key at capture time
• 903e92993e8a8ca64b5bb19919511d94fd25ce3c Revert "fix(agent-skills): restore valid skill metadata headers"
• 478adafcd1c00af8fd70deb174829292fc932fea Revert "fix(exploit-prompt): correct malformed inline code tokens"
• e24eeb0ef6d920c9193a7640fbdbc3052332bd3b Revert "refactor(static-analysis): remove exploit_intelligence stage"
• c7ad56f0d0522f4a14733e9a163dbe773dfc27e6 Revert "refactor: hoist ConfidenceLevel to guardix.confidence leaf module"
• a2ba1bdc45c47003fa5e625003e5665ed4dd17ac Revert Runtime Discipline section from checklist agent prompt
• f34f41652d5dec6eb5ff9aed114c81e1ec6378a4 Revert focused reading plan from checklist agent prompt
• 9fa05d5b05e94af1e8f50bde5297a54010f8ed6e Revert renft-overfit category guidance from checklist coordinator
• b3f4b01b29de1ced7a309336a319eb4492f327bf Review Kimi math soft stop findings
• d8d7488f4ca78f97469513111170702b39ef7e89 Review Kimi reNFT checklist findings
• eb15f610bc44bb35c11c8f7edf776485169fc0ba Review adversarial analyzer modularity
• 93ce59010912de8ee63e52f9da83d549eaefd02d Rework Codex runtime tools and caching
• 0aa7e5a6d93318e21661b4cb7bcabca2c5235d90 Rework RepoFocusPage into Linear-style project view
• ffec6e9a22632bed49e5b275239b59991c410a3a Rework exploit pipeline with execution abstractions and Cloud Run support
• 34ac84e5c9f48dad854095ededd5b515bef675bf Rework landing page: new hero, exploit showcase, micro-animations
• ac24b82413e273257bcd09a2525018b926575cfc Rework realistic to align with tdagent architecture
• 64c640812cc1042fc5fbfe6d7dc48aaa19eb8f6c Rewrite dashboard as project list with global running-audits indicator
• 187cc3f54ad3540814084f3093af456f529ec1d3 Run Medusa against harness targets
• 7f7ef23d0cc7a1ce98dc70959905a21add2ab984 Run backend/frontend CI on all PRs with fast skip
• fe775ff75ac36df58d19d91a698032bc9da62879 Run infra plan workflow on all PRs with fast skip
• 7447b0440a2c4bf6d23da05d4b7ebd454b5bd9b5 Sanitize aggregate call metrics before accumulation
• b755bb28ed18e136cf3fa25d5e9251aeaeddf0f2 Sanitize serialized call metrics in aggregate exports
• ec01a9f750b856cc6164d426450d0b270a1b64d2 Sanitize session summary IDs when listing sessions
• 315dabab89f8c73e8bedb8ddfa5bc1e6046ae276 Sanitize top-level PostHog aggregate properties before export
• abeb30376af6fba3454b29a52a6cda2ed2f22ae9 Sanitize usage counter snapshots from mutated state
• fff3bc28fa188bf1da0d24c07fac1d8dc2712f76 Score DeepSeek checklist findings
• c78e7f5054b1617fb8c8f07e523ede03d1504b74 Self-heal stale active session pointers on read
• 6bdb7f68f721e639cbe0cfff397c83f85035ab35 Serialize finding timestamps as naive UTC for Postgres
• f88afed6a44d2fe50e4d6498f09cd71d387fafba Set per-audit price to $200 (Stripe cents + marketing copy)
• 8e7644be4edb28c5762e64567383957a1d0e466a Share control-character helper across PostHog sanitizer
• c5d4611ba96264c38f3325f1555683a2e7d64c9b Show FailedAuditView when URL points to a failed audit
• 8b216741baa7ac0e22eccb01ecf939438e512adf Show baseline artifacts in CLI status when understanding is complete
• 0046a29c0b00eaa5597f5527f6f37ba16fcbc732 Show checklist run controls in status
• f33af0a09ee11c04b6261f7c8df3ab5eff1c3acf Show expected deep artifacts when deep understanding not run
• 4373d3b603a9a31cdfe9e1a77315d3d5a109f260 Show failed audit banner when navigating to failed audit URL (#138)
• 69e58dc80062d3e4213d14ddca2268b15784ab04 Show intermediate deep notes in status outputs
• d1a9eed430074e5e4854ded0d8ef4c56e20c7a5a Show present deep artifacts when understanding is incomplete
• 3dd9730c18ccfeb3541958408f8dba831bf5f970 Show question counts for incomplete deep understanding in CLI status
• eb2384c377e8f01f2d4d78b2ec11d55c4046329e Show richer diagnostics when deep-understanding command remains incomplete
• fd3de1466f80ae8304fafb6d07a5ef8a9749ec16 Simplify codex runtime finalizer recovery
• 14171b71896a61d4ddbdd90455160780fb25a68b Simplify documentation prompt wording
• f551a608361988daa8af1136a96f9dd415607991 Skip ABI resolution when Etherscan key is missing
• d78912c78dd530e5eb47009d8bbfa5e3fbf05ac9 Skip frontend pre-push tests when no frontend files change
• f1e96be2f41f9619cb79887c42724e68fed19cbc Skip infra plan and coverage comment on fork PRs
• d6d80aa17e47e54e524d99cf879270c7e3a6ba2c Skip non-object hypothesis artifacts in worker debug dump
• 0962cbe62fb1ea26619a1b21a58dc6783983af2b Spurious-FAIL guardrail: concrete-rerun for halmos witnesses
• f0420610dee8a00500f9f36d9e09e8b19f7a5f6c Stabilize CLI tests against host API key env
• 5f6f087f090d0d33b28c6638c3cab467cf9a3b15 Stabilize Guardix deployment config.
• 497d635b2b933c8d7a6e52ee4c757fdb033272b7 Stabilize admin page tab test
• 74a03469f6895157e281fa487a75a2e95d3c215d Stabilize backend coverage command in CI
• 7dfd8da7f4467c0f5936af0de07488c9a3f72847 Stabilize backend coverage comment step - pin pytest-coverage-comment to v1 - allow coverage comment to fail without breaking CI
• 55791e6c06155831390668dd5a2c153894cf25c0 Stabilize deep answers-file orchestrator test semantics
• d6bc3f501a08923db399a3122cf06fa2d5b99f32 Stabilize findings pagination test
• ed36a26a1a7443cd740c365ef1ea7f0a703604b7 Stabilize grouped audits API ordering test
• 695005bd2344f45ac80c17a7ea0b17053631b9ab Standardize subprocess text decoding across runtimes
• f7ea9e1dff2d3ccdbe3a8d4cea55141d772e3291 Start audit pipeline control-plane package
• f4a4e0dc6e4ddf3c34de0f05f080ae9997e28ba8 Start over
• 1c2f4d3f592012ce0a2a8a8d3da30fe881ec8916 Static analysis: deep understanding status repair + report fallback
• af38ffc88562d762f2c88fa9e53503334ca86060 Stop checklist runtimes after valid artifacts
• dc353a0a51467aacb5c0c62912dd9210efdc58a2 Stop metrics collection after audit finalization
• b0c416c43d82e2926774d79a55be2beccb43c3b9 Store skill-agent outputs under tob hypotheses.
• a231feaf3701fe0df122521979f8bd9d0c778667 Stream audit execution logs to RepoFocus UI
• c5b1c070056685100cce729d51be5acc69b07e7b Stream subagent runtime logs in CLI
• 4fb1179b43bd930fb4af00ebdd1a309c4890ac83 Support category-scoped checklist experiment runs
• 9a1a49713378996aeeaa55c9115e3b83257d62d5 Support hyphen aliases for deep stage override names
• a376fe8afaae73bc4ca0af83840695e1a401b95b Support markdown bullet clarification answers
• 24e488b66401469a6e36dbe2228c150df866b6a5 Support questionId key in deep clarification answer list entries
• ab016fff497a02e5ae9103c44ba8493f674fc153 Support signed integer strings in metrics and report coercion
• 72db0036bfeda2ff2824d4fb7712a6ff772d5d71 Support star and numbered clarification answer lines
• 45302053fd75a805f1f7d4ef6805ea5e8339eabb Support underscore runtime aliases in static CLI mode parsing
• 73636311f466a114a97cf20fd7bda44de4cb2b78 Support unfenced JSON snapshots in deep notes compiler
• 5a3be74f9572b5f015077c4c4f91fd3761e79e49 Surface baseline understanding validation errors in audit status and CLI
• f4613cac3c5f93360d9828eeb6c75618ffe798df Surface deep artifact validation errors in audit status
• d4648ceb0f0a74836f57bb0c1559a107d4be4f00 Switch Astro blog pages to dark theme styling
• 900444b2f12fc2bbe97068c9158d46ad60f687b6 Switch default validators to claude/grok/gemini
• f5ed2a8ecb60cef8476cdd4ae981cfd9bebc9e32 Tail live logs for active checklist status
• dd6a5dcd31a7eeb130859a08cc7f39d71a174990 Tee agent logs with audit id prefix
• 7f30c5611f0d3fa32e49486b08eb0ceb4e9de799 Tell checklist agents to close on evidence
• e72061fb39a54cb81504efc460ee35e089f6bb63 Test Cloud Run job env overrides
• 0653c23d2535f6056984caa52add23f4eef5c8d0 Tests: update fixtures to use RuntimeConfig
• 4849fd7a122587369bc6849843c54a6bac16906a Throttle OpenRouter checklist live output
• 0bf68394382d9627973a94aa3796e7971aad49df Tighten deployment and exploit prompts for iterative exploitation
• d0f7e3abdd1da757ddcfa4296f907968264c7c12 Tighten subagent runtime typing
• 159bd53a4a016313e9a9cfaa3137ffd7a1d47737 Tone down prompt language - less shouting, more professional
• 728c9dd01e319a75e26a2cd3f264b1056bd8391a Track OpenRouter provider cost for checklist runs
• a1a3982c422be365f3ebe9d7d44e8cd5c28d3c52 Track OpenRouter stream activity
• af779f01f4e006943df74c472079e178d198f676 Track adversarial investigator progress
• 5e09130f6a2bc41c578243893e53e315306ac3ac Track streamed model activity in runtime heartbeat
• 90e6a4fc0d1821d314d91856d8170f296f525a32 Treat all-invalid session lists as empty in CLI output
• 20aa7c124a905393af851d4967158555dfca5394 Treat control-char disable flags as PostHog opt-out
• 9a234d8cfa5dfdeef6ac28cd188913f262007606 Treat empty forced checklist finalizers as incomplete
• 60cb88e5906e26cea4a83fffae0488f2b2c8e73c Treat non-boolean include-tools flags as disabled
• 760ec00d225a1a168073ef19448bbfaa60de89a8 Treat only datetime finished_at as terminal metrics state
• 44c77a45e9b46cd5d5e52e75c52b1175edbf19d8 Treat validated tool finalizer output as success
• 8f04a10a8ca34b5d0ca696c84f258ae46f1ad831 Trim and validate PostHog environment configuration values
• 21c121547a24d2ae22c48d3f519a8a6931ac86ed Trim target repository component in PostHog distinct IDs
• 8da1d160739e349e4c7d1f9122375c6878f80c53 Tune audit polling intervals and add coverage
• aad641f087d8d89bf1bb68096dab42ab0b683019 UI refresh: severity ring dots, file chips, dark theme tokens
• 415451ab9587d88fc5d51741691c5309418caa1a Unify Python CLI modules
• d99f2fb4ffbb92cfc2149a32afc388c56d834a2c Unify deep artifact filename set via shared schema constant
• 5d9394aaa5ff3da6d336aa060d92fa39999331ea Update .gitignore for experiment artifacts and Foundry builds
• 027c82113b9a28c289a192514ed829d58a20bdf0 Update DeepSeek BendDAO checklist run
• 9557e2d2abde0eb868f42e8bb2654a02e5b2674b Update adversarial analyzer progress tracker
• 79dc1965648c2e7dbcfbde594297812d09a39385 Update adversarial validation blockers
• 8cfcbe41ef243e64d0f73b86c8c1c6827ce2443d Update adversarial validation status
• e6ec79cc63603f8c1f43fd7744372efbee41ef09 Update audit detail page test title
• e2927d620d00548b15b8b9a106ddcd3a03b935fe Update checklist investi...