Skip to content

Releases: sickn33/agentic-awesome-skills

v13.13.0 "Codex Profiles, Deep Research, and Agent Reliability Gates"

Choose a tag to compare

@sickn33 sickn33 released this 08 Jul 05:44

[13.13.0] - 2026-07-08 - "Codex Profiles, Deep Research, and Agent Reliability Gates"

Codex profile isolation, deep research orchestration, pre-ship verification, agent networking, tree-ring memory, and synchronized public metadata for the 1,935+ skill catalog.

Start here:

This release ships the July 8 maintenance batch: five merged community PRs, a source-backed Codex Profiles follow-up for issue #789, generated registry/plugin mirror sync, and refreshed public SEO/LLM metadata for the 1,935+ skill catalog.

Added

  • Added go-in-depth, a self-authored deep research workflow skill with multi-agent fan-out, evidence synthesis, and verification guidance (PR #786).
  • Added routerbase-model-gateway, a community skill for RouterBase model gateway setup, routing, observability, and fallback operations (PR #787).
  • Added pre-ship-gate, a community pre-release verification skill for catching silent failures across migrations, feature flags, build caches, release pointers, staged rollouts, and environment drift (PR #788).
  • Added pilot-protocol, a community agent-networking skill for cross-agent coordination, handoff, and message protocol reliability (PR #790).
  • Added tree-ring-memory, a community memory governance skill for durable, append-only learning records and reviewable decision history (PR #791).
  • Added codex-profiles, a critical-risk Codex CLI/Desktop profile isolation skill sourced from Ducksss/codex-profiles, with guidance for separate CODEX_HOME profiles and explicit warnings not to copy or expose auth tokens (issue #789).

Changed

  • Refreshed generated registry artifacts, plugin compatibility metadata, plugin mirrors, package metadata, public docs, sitemap, web catalog assets, llms.txt, social-card metadata, and SEO fallbacks for the 1,935+ skill catalog.
  • Added or repaired README source credits for RouterBase, pre-ship-gate, pilot-protocol, tree-ring-memory, and codex-profiles source material.

Validation

  • Verified and merged PR #786, PR #787, PR #788, PR #790, and PR #791 after maintainer review and targeted fixes.
  • Closed issue #789 after adding the source-backed codex-profiles skill and regenerated catalog surfaces.
  • Ran npm run check:readme-credits, npm run validate, npm run security:docs, npm run app:build, and env npm_config_cache=/private/tmp/aas-npm-cache npm run test during the maintainer sweep.
  • Ran npm run sync:repo-state and refreshed public SEO/LLM surfaces before release preparation.

Credits

v13.12.0 "Agent Workflow Skills, Social Publishing, and Loki Dependency Refresh"

Choose a tag to compare

@sickn33 sickn33 released this 07 Jul 07:04

[13.12.0] - 2026-07-07 - "Agent Workflow Skills, Social Publishing, and Loki Dependency Refresh"

Agent-orchestration workflows, social publishing coverage, LinkedIn writing support, and synchronized dependency maintenance for the 1,929+ skill catalog.

Start here:

This release ships the July 7 maintenance batch: three merged PRs, a source-backed Taisly follow-up skill, 26 imported David Ondrej agent workflow skills, a LinkedIn post-writing skill with bundled hook references, a canonical Loki example dependency refresh, and regenerated registry/public catalog state for the 1,929+ skill catalog.

Added

  • Added linkedin-post-writer, a community LinkedIn post drafting skill adapted from sergebulaev/linkedin-skills, including bundled reference material for 16 hook formulas mapped to engagement goals (PR #785).
  • Added taisly-social-media-posting, a critical-risk social video publishing workflow skill backed by taisly/agent, with explicit approval gates for publishing, scheduling, account-linking, and metadata updates (PR #783 plus maintainer source-backed skill follow-up).
  • Imported 26 non-colliding MIT-licensed skills from davidondrej/skills:
    • agent-self-scheduling
    • anti-sleep
    • brain-to-docs
    • browser-harness
    • cmux
    • codex-subagent
    • cyber-audit
    • deepapi
    • delegating-to-agents
    • distribute-skill-to-all-agents
    • effective-agent-skills
    • fable-safe-prompt
    • folder-specific-claude-and-agents-md
    • goal-loop
    • interview-style-doc-building
    • markdown-rendering
    • pi-custom-model
    • pi-web-search
    • push-skill-to-github
    • read-all-adrs
    • research-prompt
    • run-deep-swe
    • setup-help
    • short
    • vps-server-management
    • youtube-transcript
  • Added README source credits for sergebulaev/linkedin-skills, taisly/agent, and davidondrej/skills.

Changed

  • Updated the Loki generated todo backend example dependency better-sqlite3 from ^12.10.1 to ^12.11.1 in the canonical skill source, then synchronized plugin mirror state from the maintained source (PR #784).
  • Normalized imported David Ondrej skills with AAS frontmatter, concise descriptions, source/license metadata, ## When to Use sections, and safety-oriented limitations.
  • Refreshed generated registry artifacts, plugin compatibility metadata, plugin mirrors, public web skill assets, sitemap, package description, and README/docs counters for the 1,929+ skill catalog.

Validation

  • Verified and merged PR #783, PR #784, and PR #785 after required GitHub checks passed.
  • Repaired PR #784 from a generated mirror-only dependency bump into a source-only canonical Loki example dependency update before merge.
  • Ran npm run validate, npm run security:docs, npm run check:warning-budget, and npm run check:readme-credits -- --base origin/main --head HEAD during the maintainer import pass.
  • Ran npm run sync:repo-state after merging PRs and importing new skills.

Credits

v13.11.0 "Ledger Skills, WGM, and Loki Dependency Maintenance"

Choose a tag to compare

@sickn33 sickn33 released this 05 Jul 16:07

[13.11.0] - 2026-07-05 - "Ledger Skills, WGM, and Loki Dependency Maintenance"

Notion-backed ledger skills, a governed build-loop methodology, and synchronized dependency maintenance for the 1,901+ skill catalog.

Start here:

This release ships the July 5 maintenance batch: three source-backed skills, README upstream credits, a maintainer-side Snyk dependency repair for the Loki generated todo example, and refreshed registry/public catalog state for the 1,901+ skill catalog.

Added

  • Added time-ledger, a critical-risk Notion skill that parses natural-language time reports into a user-owned Notion database while marking uncertain entries To-confirm instead of guessing (PR #777).
  • Added trading-ledger, a critical-risk Notion trading journal skill that records thesis, plan, emotion, fills, closes, and reviews while explicitly avoiding financial advice and market-data lookups (PR #778).
  • Added wgm, a safe meta skill for governed build loops with triage, alignment, planning, deterministic backpressure, holdout-scenario judging, and handoff audits (PR #782).
  • Added README upstream credits for cruisekkk/time-ledger, cruisekkk/trading-ledger, and agent-frontier/wgm.

Changed

  • Updated the Loki generated todo backend example dependency better-sqlite3 from ^12.10.0 to ^12.10.1 in both the canonical skill source and Claude plugin mirror, superseding the generated-only Snyk PR #779 with a source-synchronized maintainer patch.
  • Refreshed generated registry artifacts, plugin compatibility metadata, plugin mirrors, public web skill assets, sitemap, package description, and README/docs counters for the 1,901+ skill catalog.

Validation

  • Verified and merged PR #777, PR #778, and PR #782 after required GitHub checks passed.
  • Ran npm run security:docs and git diff --check for the Loki dependency maintenance patch.
  • Ran npm run validate:references, npm run check:readme-credits -- --base origin/main --head HEAD, and git diff --check during the maintainer fixes for PR #778 and PR #782.
  • Ran npm run sync:repo-state before release preparation.
  • Ran the release prepare suite for v13.11.0, including reference validation, release-state sync, tests, web-app install, web-app build, and package dry run.
  • Ran cd apps/web-app && npm run verify:seo.

Credits

v13.10.0 "Context, WordPress, and PR Intake Hardening"

Choose a tag to compare

@sickn33 sickn33 released this 04 Jul 15:05

[13.10.0] - 2026-07-04 - "Context, WordPress, and PR Intake Hardening"

Community skill intake, ASO research guidance, pull-request CI hardening, and catalog sync for the 1,898+ skill catalog.

Start here:

This release ships the July 4 maintenance batch: two new source-backed skills, one ASO documentation improvement, repaired pull-request intake checks for fork contributions, generated registry and plugin mirror sync, and refreshed public catalog surfaces for the 1,898+ skill catalog.

Added

  • Added wp-site-health-auditor, a critical-risk WordPress Site Health skill that turns Site Health reports into risk-tiered, backup-first remediation plans with WP-CLI, PHP, .htaccess, cache, HTTPS, and rollback guidance (PR #775).
  • Added context-kit, a critical-risk personal-context hygiene skill for safely evaluating, adapting, and installing Context Kit-style Personal Context Artifact workflows without leaking secrets or private Markdown context (PR #776).
  • Added the JDDavenport/context-kit source reference to README Community Contributors so the new external-source-backed skill has explicit attribution.

Changed

  • Updated app-store-optimization with optional AppNiche external research tools for directional iOS revenue benchmarks and keyword opportunity checks, with retrieval-date and approximation guidance (PR #774).
  • Refreshed generated registry artifacts, plugin compatibility metadata, plugin mirrors, public web skill assets, sitemap, package description, and README/docs counters for the 1,898+ skill catalog.

Fixed

  • Fixed PR intake CI by installing npm dependencies before tools/scripts/pr_preflight.cjs runs in the pr-policy job.
  • Fixed tools/scripts/pr_preflight.cjs so GitHub Actions absolute GITHUB_EVENT_PATH values are read correctly, with a regression test for absolute event paths.

Validation

  • Verified and merged PR #774, PR #775, and PR #776 after required GitHub checks passed.
  • Ran npm run test after the PR intake fix.
  • Ran npm run sync:repo-state.
  • Ran npm run validate and npm run security:docs during the Context Kit repair pass.
  • Ran npm run check:readme-credits -- --base origin/main --head HEAD for the Context Kit source-credit gate.

Credits

v13.9.0

Choose a tag to compare

@sickn33 sickn33 released this 03 Jul 08:30

[13.9.0] - 2026-07-03 - "WorkorAI, Autohand, and Web Dependency Refresh"

Community skill intake, host documentation, dependency maintenance, and catalog sync for the 1,896+ skill catalog.

Start here:

This release completes the July 3 maintenance batch after v13.8.0: the WorkorAI community skill, Autohand Code install notes, a web-app Supabase dependency refresh, generated registry and public catalog updates, and refreshed SEO/public surfaces for the 1,896+ skill catalog.

Added

  • Added workorai, a critical-risk WorkorAI MCP skill for candidate job search/application flows and employer job lifecycle, candidate discovery, invitations, and applicant review workflows (PR #773).
  • Added Autohand Code as a documented host in the README tool matrix, badges, intro copy, and install FAQ using the generic --path ~/.autohand/skills and --path .autohand/skills installer flow (manual integration of PR #772).

Changed

  • Updated README metadata sync tooling so Autohand Code remains part of generated README copy instead of being removed by future maintainer sync runs.
  • Updated the web app @supabase/supabase-js dependency manifest to ^2.110.0, matching the current lockfile resolution and superseding the stale Snyk 2.108.1 upgrade PR (#770).
  • Refreshed generated registry artifacts, plugin compatibility metadata, catalog data, sitemap, public web skill assets, package description, and README counters for the 1,896+ skill catalog.

Validation

  • Verified and merged PR #773 after GitHub reported it mergeable and external security checks were successful.
  • Reviewed PR #772 and integrated its README changes against the current 13.8.0/1,896+ public copy instead of applying its stale 13.7.0 patch.
  • Reviewed PR #770 and applied the newer resolved Supabase dependency state rather than downgrading from the current lockfile to 2.108.1.
  • Ran npm install @supabase/supabase-js@^2.110.0 --package-lock-only in apps/web-app with 0 vulnerabilities.
  • Ran npm run sync:repo-state.
  • Ran the release prepare suite for v13.9.0, including reference validation, release-state sync, tests, web-app install, web-app build, and package dry run.
  • Ran cd apps/web-app && npm run verify:seo.

Credits

v13.8.0

Choose a tag to compare

@sickn33 sickn33 released this 03 Jul 08:12

[13.8.0] - 2026-07-03 - "Code Polish and Catalog Sync"

Community skill intake and maintainer sync for the 1,895+ skill catalog.

Start here:

This release packages the July 3 maintenance pass: one fully-checked community skill PR, generated registry and plugin mirror sync, public catalog counters, and a clean SEO/public-surface refresh for the 1,895+ skill catalog.

Added

  • Added code-polish, a constraint-based cleanup skill for professionalizing code comments, removing redundant or stale comment noise, and keeping behavior changes out of scope (PR #771).

Changed

  • Refreshed generated registry artifacts, plugin mirrors, catalog data, plugin compatibility metadata, public docs, sitemap, package description, and README counters for the 1,895+ skill catalog.

Validation

  • Verified and merged PR #771 after required GitHub checks passed.
  • Ran npm run sync:repo-state.
  • Ran npm_config_cache=/private/tmp/aas-npm-cache npm audit --audit-level=moderate with 0 vulnerabilities.
  • Ran the release prepare suite for v13.8.0, including reference validation, release-state sync, tests, web-app install, web-app build, and package dry run.
  • Ran cd apps/web-app && npm run verify:seo.

Credits

v13.7.0 "Security Hardening and Community Intake"

Choose a tag to compare

@sickn33 sickn33 released this 02 Jul 06:09

[13.7.0] - 2026-07-02 - "Security Hardening and Community Intake"

Maintainer security sweep, PR maintenance, and catalog sync for the 1,894+ skill catalog.

Start here:

This release packages the July 2 maintenance batch: three community PRs, hardening for the reviewed security findings CSV, generated registry and plugin mirror sync, and a catalog quality fix for product-risk skill categorization.

Added

  • Added mdpr-skill, a Codex-assisted MDPR presentation review skill for semantic hints, visual checks, theme candidates, and deterministic renderer boundaries (PR #767).
  • Added dispatch, a multi-CLI delegation skill for routing work from Claude Code to Codex, Antigravity, and Gemini agents (PR #768).
  • Added before-you-build, a product-risk review skill for checking demand, alternatives, switching costs, channels, and validation steps before coding (PR #769).

Changed

  • Hardened the skill-review workflow so Tessl credentials are only exposed to trusted, pinned setup steps while PR content is reviewed through repository-owned trusted scripts.
  • Hardened hugging-face-model-trainer GGUF conversion by validating Hugging Face repo IDs and making trust_remote_code an explicit TRUST_REMOTE_CODE opt-in.
  • Hardened weaviate connection handling so provider API keys are forwarded only through an explicit WEAVIATE_PROVIDER_KEYS allowlist.
  • Reclassified sql-sentinel and gh-image as critical and removed them from generated Codex/Claude plugin mirrors until their upstream execution and browser-session guidance is reviewed for plugin distribution.
  • Reclassified riffkit as critical because it can use a session token and paid generation endpoint.
  • Fixed symlink handling in youtube-notetaker, malformed numeric token parsing in cron-doctor, and the misplaced Hive helper import in monte-carlo-push-ingestion.
  • Updated catalog generation so explicit product/business skill frontmatter is kept in the business catalog lane instead of being misclassified as security because of generic risk wording.
  • Refreshed generated registry artifacts, plugin mirrors, catalog data, plugin compatibility metadata, public docs, sitemap, llms.txt, package description, and README counters for the 1,894+ skill catalog.

Validation

  • Verified and merged PR #767, PR #768, and PR #769 after required GitHub checks passed.
  • Ran npm run sync:repo-state.
  • Ran npm run security:docs.
  • Ran npm_config_cache=/private/tmp/aas-npm-cache npm audit --json with 0 vulnerabilities.
  • Ran npm_config_cache=/private/tmp/aas-npm-cache npm run test.
  • Ran node tools/scripts/tests/build_catalog_bundles.test.js.
  • Ran npm run app:test.
  • Ran npm run app:build.
  • Ran cd apps/web-app && npm run verify:seo.

Credits

v13.6.1 "Source Catalog, Riffkit, and Tessl Review"

Choose a tag to compare

@sickn33 sickn33 released this 01 Jul 11:46

[13.6.1] - 2026-07-01 - "LLMS SEO Metadata Refresh"

Patch release for hosted catalog SEO metadata after the 1,891+ skill refresh.

Fixed

  • Refreshed apps/web-app/public/llms.txt so the GitHub Pages SEO verification gate matches the current V13.6.1 release metadata and 1,891+ skill count.

Validation

  • Ran npm run release:preflight.
  • Ran npm run security:docs.
  • Verified GitHub Pages deploy after publishing the patch release.

Included Source Sweep From v13.6.0

[13.6.0] - 2026-07-01 - "Source Catalog, Riffkit, and Tessl Review"

Maintainer source sweep, community intake, and workflow maintenance for the 1,891+ skill catalog.

Start here:

This release packages the July 1 maintainer batch: a source-repository refresh across official, community, inspiration, and additional source collections; 191 newly-added canonical SKILL.md entries; 38 refreshed canonical SKILL.md entries; editorial pruning and normalization; plugin mirror sync; two community PRs; and pull-request workflow maintenance. No canonical skill entries were removed.

Added

  • Added 191 canonical SKILL.md entries from the source sweep, including the families below.
  • Added experimentation, analytics, growth, and product-marketing skills: ab-testing, analytics, alternatives-pages, co-marketing, community-building, competitor-profiling, competitor-tracking, cro, customer-research, developer-advocacy, developer-audience-context, developer-churn, developer-listening, developer-newsletter, developer-onboarding, developer-sandbox, developer-seo, developer-signup-flow, devrel-content, docs-as-marketing, free-tier-strategy, github-presence, marketing-plan, offers, onboarding, open-source-marketing, power-user-cultivation, pricing, product-marketing, public-relations, sdk-dx, and usage-based-pricing.
  • Added API, platform, cloud, and database skills: api-analyzer, api-and-interface-design, api-designer, api-integration, api-onboarding, api-sdk-generator, applicationinsights-web-ts, aws-agentic-ai, aws-cdk-development, aws-cost-operations, aws-mcp-setup, aws-serverless-eda, aws-sst-development, azure-ai-language-conversations-py, azure-servicebus-rust, azure-storage-queue-rust, graphql-schema, neon-ai-gateway, neon-functions, neon-object-storage, neon-postgres-branches, neon-postgres-egress-optimizer, openapi-spec-generator, postgresql-cli, postman-collection-generator, postman-newman-automation, postman-openapi-converter, rclone-cli, redis-cli, supabase, and supabase-postgres-best-practices.
  • Added Expo, mobile, native-app, and SwiftUI-adjacent skills: add-app-clip, eas-update-insights, expo-brownfield, expo-examples, expo-module, expo-observe, expo-ui, and use-dom.
  • Added AI, agent, context-engineering, Hugging Face, Gemini, and model-building skills: agent-memory, anti-deception, context-engineering, doubt-driven-development, faf-context, faf-go, gemini-interactions-api, gemini-live-api-dev, gemini-omni-flash-api, hf-mcp, hf-mem, huggingface-best, huggingface-local-models, huggingface-lora-space-builder, huggingface-spaces, huggingface-tool-builder, huggingface-zerogpu, privacy-mask, train-sentence-transformers, trl-training, and tune-monitor.
  • Added frontend, UI, UX, design, and presentation skills: design-philosophy, design-spatial, design-system, design-thinking, design-ux, formik-patterns, frontend-architecture, frontend-data-contracts, frontend-observability, frontend-optimistic-mutations, frontend-seo, frontend-slides-frontend-slides, frontend-ui-engineering, styleseed-design-review, ui-lint, ui-motion, ui-score, ui-skills-root, and ui-update.
  • Added testing and QA skills: appium-skill, cucumber-skill, cypress-skill, jest-skill, junit-5-skill, newman-cicd-integration, puppeteer-skill, pytest-skill, robot-framework-skill, selenium-skill, test-framework-migration-skill, testng-skill, vitest-skill, webdriverio-skill, wjttc-builder, and wjttc-tester.
  • Added code-quality, review, release, and maintainer workflow skills: accint-commitments, accint-frames, automated-triage, brooks-audit, brooks-debt, brooks-harness, brooks-review, brooks-sweep, brooks-test, bug-hunt-swarm, changelog-updates, ci-cd-and-automation, code-review-and-quality, code-showcase-core-components, code-showcase-react-ui-patterns, code-showcase-systematic-debugging, code-showcase-testing-patterns, code-simplification, debugging-and-error-recovery, debugging-code, deprecation-and-migration, documentation-and-adrs, git-workflow-and-versioning, implement, incremental-implementation, logic-diff, logic-explain, logic-fix-all, logic-locate, logic-review, loopy, planning-and-task-breakdown, repo-maintainer, resolving-merge-conflicts, review-and-simplify-changes, review-swarm, security-and-hardening, shipping-and-launch, source-driven-development, spec-driven-development, unslop-commit, unslop-file, and unslop-review.
  • Added docs, content, tutorial, data-observability, finance, and specialized utility skills: dev-to-hashnode, hugo-to-markdown, hyperexecute-skill, idea-refine, longbridge-content, longbridge-fundamentals, longbridge-market-data, monte-carlo-analyze-root-cause, monte-carlo-asset-health, monte-carlo-monitoring-advisor, monte-carlo-performance-diagnosis, monte-carlo-remediation, monte-carlo-storage-cost-analysis, observability-and-instrumentation, odw, performance-optimization, polis-protocol-a-self-optimizing-city-of-agents, smartui-skill, technical-tutorials, and update-swiftui-apis.
  • Added riffkit, a hosted short-form video skill for generating user-confirmed TikTok/UGC ad riffs from a source video, product, character, and language (PR #765).
  • Added Vexilo to the README comparison resources as a visual field guide for Claude Code primitives and workflows (PR #766).

Changed

  • Migrated the skill-review pull-request workflow from the deprecated Tessl Skill Review action to Tessl Review via tesslio/setup-tessl and a repository-owned changed-skill runner.
  • Paused the maintainer-only /apply-optimize shortcut and its manual runner until the new tessl review fix output contract is wired safely.
  • Refreshed 38 existing canonical SKILL.md entries: accint-solve, baseline-ui, building-native-ui, claimable-postgres, expo-api-routes, expo-cicd-workflows, expo-deployment, expo-dev-client, expo-tailwind-setup, fixing-accessibility, fixing-metadata, fixing-motion-performance, gemini-api-dev, hugging-face-cli, hugging-face-community-evals, hugging-face-dataset-viewer, hugging-face-gradio, hugging-face-model-trainer, hugging-face-paper-publisher, hugging-face-papers, hugging-face-trackio, hugging-face-vision-trainer, native-data-fetching, neon-postgres, swiftui-expert-skill, transformers-js, ui-a11y, ui-component, ui-page, ui-pattern, ui-review, ui-setup, ui-tokens, upgrading-expo, ux-audit, ux-copy, ux-feedback, and ux-flow.
  • Added or refreshed bundled references, agents, evals, and scripts for source-backed skills, including Expo agents/references, Hugging Face model and Spaces references, Postman/Newman workflows, RClone command/provider references, Redis command references, SwiftUI trace tooling, UI/design references, and training scripts for sentence-transformers and Gemini Omni Flash workflows.
  • Normalized imported skill metadata, limitations, safety notes, source provenance, license fields, bundled references, scripts, and plugin compatibility surfaces across the source-sweep batch.
  • Refreshed generated registry artifacts, plugin mirrors, catalog data, plugin compatibility metadata, public docs, sitemap, package description, llms.txt, and README counters for the 1,891+ skill catalog.

Validation

  • Verified PR #765 checks for Devin Review, Socket, Snyk license, and Snyk security before merging.
  • Verified PR #766 checks for Devin Review, Socket, Snyk license, and Snyk security before merging.
  • Ran npm run validate.
  • Ran npm run validate:references.
  • Ran npm run security:docs.
  • Ran env npm_config_cache=/private/tmp/aas-npm-cache npm run test:local.
  • Ran npm run check:readme-credits.
  • Ran the maintainer source-credit gate and git diff --cached --check before publishing the source refresh.

Credits

v13.6.0

Choose a tag to compare

@sickn33 sickn33 released this 01 Jul 11:38

[13.6.0] - 2026-07-01 - "Source Catalog, Riffkit, and Tessl Review"

Maintainer source sweep, community intake, and workflow maintenance for the 1,891+ skill catalog.

Start here:

This release packages the July 1 maintainer batch: a source-repository refresh across official, community, inspiration, and additional source collections; 191 newly-added canonical SKILL.md entries; 38 refreshed canonical SKILL.md entries; editorial pruning and normalization; plugin mirror sync; two community PRs; and pull-request workflow maintenance. No canonical skill entries were removed.

Added

  • Added 191 canonical SKILL.md entries from the source sweep, including the families below.
  • Added experimentation, analytics, growth, and product-marketing skills: ab-testing, analytics, alternatives-pages, co-marketing, community-building, competitor-profiling, competitor-tracking, cro, customer-research, developer-advocacy, developer-audience-context, developer-churn, developer-listening, developer-newsletter, developer-onboarding, developer-sandbox, developer-seo, developer-signup-flow, devrel-content, docs-as-marketing, free-tier-strategy, github-presence, marketing-plan, offers, onboarding, open-source-marketing, power-user-cultivation, pricing, product-marketing, public-relations, sdk-dx, and usage-based-pricing.
  • Added API, platform, cloud, and database skills: api-analyzer, api-and-interface-design, api-designer, api-integration, api-onboarding, api-sdk-generator, applicationinsights-web-ts, aws-agentic-ai, aws-cdk-development, aws-cost-operations, aws-mcp-setup, aws-serverless-eda, aws-sst-development, azure-ai-language-conversations-py, azure-servicebus-rust, azure-storage-queue-rust, graphql-schema, neon-ai-gateway, neon-functions, neon-object-storage, neon-postgres-branches, neon-postgres-egress-optimizer, openapi-spec-generator, postgresql-cli, postman-collection-generator, postman-newman-automation, postman-openapi-converter, rclone-cli, redis-cli, supabase, and supabase-postgres-best-practices.
  • Added Expo, mobile, native-app, and SwiftUI-adjacent skills: add-app-clip, eas-update-insights, expo-brownfield, expo-examples, expo-module, expo-observe, expo-ui, and use-dom.
  • Added AI, agent, context-engineering, Hugging Face, Gemini, and model-building skills: agent-memory, anti-deception, context-engineering, doubt-driven-development, faf-context, faf-go, gemini-interactions-api, gemini-live-api-dev, gemini-omni-flash-api, hf-mcp, hf-mem, huggingface-best, huggingface-local-models, huggingface-lora-space-builder, huggingface-spaces, huggingface-tool-builder, huggingface-zerogpu, privacy-mask, train-sentence-transformers, trl-training, and tune-monitor.
  • Added frontend, UI, UX, design, and presentation skills: design-philosophy, design-spatial, design-system, design-thinking, design-ux, formik-patterns, frontend-architecture, frontend-data-contracts, frontend-observability, frontend-optimistic-mutations, frontend-seo, frontend-slides-frontend-slides, frontend-ui-engineering, styleseed-design-review, ui-lint, ui-motion, ui-score, ui-skills-root, and ui-update.
  • Added testing and QA skills: appium-skill, cucumber-skill, cypress-skill, jest-skill, junit-5-skill, newman-cicd-integration, puppeteer-skill, pytest-skill, robot-framework-skill, selenium-skill, test-framework-migration-skill, testng-skill, vitest-skill, webdriverio-skill, wjttc-builder, and wjttc-tester.
  • Added code-quality, review, release, and maintainer workflow skills: accint-commitments, accint-frames, automated-triage, brooks-audit, brooks-debt, brooks-harness, brooks-review, brooks-sweep, brooks-test, bug-hunt-swarm, changelog-updates, ci-cd-and-automation, code-review-and-quality, code-showcase-core-components, code-showcase-react-ui-patterns, code-showcase-systematic-debugging, code-showcase-testing-patterns, code-simplification, debugging-and-error-recovery, debugging-code, deprecation-and-migration, documentation-and-adrs, git-workflow-and-versioning, implement, incremental-implementation, logic-diff, logic-explain, logic-fix-all, logic-locate, logic-review, loopy, planning-and-task-breakdown, repo-maintainer, resolving-merge-conflicts, review-and-simplify-changes, review-swarm, security-and-hardening, shipping-and-launch, source-driven-development, spec-driven-development, unslop-commit, unslop-file, and unslop-review.
  • Added docs, content, tutorial, data-observability, finance, and specialized utility skills: dev-to-hashnode, hugo-to-markdown, hyperexecute-skill, idea-refine, longbridge-content, longbridge-fundamentals, longbridge-market-data, monte-carlo-analyze-root-cause, monte-carlo-asset-health, monte-carlo-monitoring-advisor, monte-carlo-performance-diagnosis, monte-carlo-remediation, monte-carlo-storage-cost-analysis, observability-and-instrumentation, odw, performance-optimization, polis-protocol-a-self-optimizing-city-of-agents, smartui-skill, technical-tutorials, and update-swiftui-apis.
  • Added riffkit, a hosted short-form video skill for generating user-confirmed TikTok/UGC ad riffs from a source video, product, character, and language (PR #765).
  • Added Vexilo to the README comparison resources as a visual field guide for Claude Code primitives and workflows (PR #766).

Changed

  • Migrated the skill-review pull-request workflow from the deprecated Tessl Skill Review action to Tessl Review via tesslio/setup-tessl and a repository-owned changed-skill runner.
  • Paused the maintainer-only /apply-optimize shortcut and its manual runner until the new tessl review fix output contract is wired safely.
  • Refreshed 38 existing canonical SKILL.md entries: accint-solve, baseline-ui, building-native-ui, claimable-postgres, expo-api-routes, expo-cicd-workflows, expo-deployment, expo-dev-client, expo-tailwind-setup, fixing-accessibility, fixing-metadata, fixing-motion-performance, gemini-api-dev, hugging-face-cli, hugging-face-community-evals, hugging-face-dataset-viewer, hugging-face-gradio, hugging-face-model-trainer, hugging-face-paper-publisher, hugging-face-papers, hugging-face-trackio, hugging-face-vision-trainer, native-data-fetching, neon-postgres, swiftui-expert-skill, transformers-js, ui-a11y, ui-component, ui-page, ui-pattern, ui-review, ui-setup, ui-tokens, upgrading-expo, ux-audit, ux-copy, ux-feedback, and ux-flow.
  • Added or refreshed bundled references, agents, evals, and scripts for source-backed skills, including Expo agents/references, Hugging Face model and Spaces references, Postman/Newman workflows, RClone command/provider references, Redis command references, SwiftUI trace tooling, UI/design references, and training scripts for sentence-transformers and Gemini Omni Flash workflows.
  • Normalized imported skill metadata, limitations, safety notes, source provenance, license fields, bundled references, scripts, and plugin compatibility surfaces across the source-sweep batch.
  • Refreshed generated registry artifacts, plugin mirrors, catalog data, plugin compatibility metadata, public docs, sitemap, package description, llms.txt, and README counters for the 1,891+ skill catalog.

Validation

  • Verified PR #765 checks for Devin Review, Socket, Snyk license, and Snyk security before merging.
  • Verified PR #766 checks for Devin Review, Socket, Snyk license, and Snyk security before merging.
  • Ran npm run validate.
  • Ran npm run validate:references.
  • Ran npm run security:docs.
  • Ran env npm_config_cache=/private/tmp/aas-npm-cache npm run test:local.
  • Ran npm run check:readme-credits.
  • Ran the maintainer source-credit gate and git diff --cached --check before publishing the source refresh.

Credits

v13.5.0 "Release Readiness, Weaviate, and Browser Runtime Gates"

Choose a tag to compare

@sickn33 sickn33 released this 29 Jun 09:50

[13.5.0] - 2026-06-29 - "Release Readiness, Weaviate, and Browser Runtime Gates"

Community skill intake and curated-source release for the 1,700+ skill catalog.

Start here:

This release packages the June 29 maintainer batch: one community PR, six curated external skill imports, source-credit updates, generated registry and plugin mirror sync, and public count refresh for the 1,700+ skill catalog.

Added

  • Added re-create, a controlled erasure and rebuild protocol for cases where a file or module must be rewritten from scratch after explicitly preserving public interfaces, working behavior, and blast-radius knowledge (PR #760).
  • Added pre-release-review from chaunsin/agent-skills, a read-only production release-readiness audit for migrations, config, secrets, deploy ordering, rollback risk, and launch blockers.
  • Added drizzle-migration-conflict from chaunsin/agent-skills, a focused Drizzle Kit migration-conflict diagnosis, repair, and prevention workflow with references and a database-free helper script.
  • Added weaviate from weaviate/agent-skills, an official Weaviate operations skill for collection inspection, semantic/hybrid/keyword search, data imports, and query-agent workflows.
  • Added weaviate-cookbooks from weaviate/agent-skills, an official cookbook index for Weaviate RAG, agentic RAG, multimodal PDF search, data explorer, chatbot, frontend, and async-client blueprints.
  • Added frontend-lighthouse from stareezy-1/frontend-architecture-skill, a portable Lighthouse CI performance gate for Core Web Vitals budgets, category floors, median runs, and CI artifacts.
  • Added browser-testing-with-devtools from addyosmani/agent-skills, a Chrome DevTools MCP browser-verification skill for live DOM, console, network, screenshot, accessibility, and performance evidence.

Changed

  • Added README source credits for chaunsin/agent-skills, weaviate/agent-skills, stareezy-1/frontend-architecture-skill, and addyosmani/agent-skills.
  • Normalized imported skill metadata with repository categories, risk labels, source provenance, dates, authors, tags, tools, license fields, When to Use sections, and explicit Limitations.
  • Rewrote Weaviate cookbook install snippets that piped remote install scripts into a shell so the docs-security check keeps download, inspection, and execution as separate steps.
  • Refreshed generated registry artifacts, plugin mirrors, catalog data, plugin compatibility metadata, package description, README counters, and public docs for the 1,700+ skill catalog.

Validation

  • Verified PR #760 checks for source validation, artifact preview, review, CodeQL, Dependency Review, Socket, Snyk, and PR policy before merging.
  • Ran npm run validate.
  • Ran npm run security:docs after normalizing the imported Weaviate install snippets.
  • Ran npm run check:warning-budget.
  • Ran npm run chain, including validation, plugin compatibility sync, index generation, bundle sync, and metadata sync.
  • Ran npm run catalog.

Credits