Skip to content

Releases

MESA v0.9.3 — ingestion and parser improvements

Choose a tag to compare

@Yasou13 Yasou13 released this 17 Nov 03:32
531209b

MESA 0.9.3 improves bulk ingestion and legal parsing, updates the default embedding model, and fixes session, projection, and retrieval edge cases. Existing V4 HTTP routes remain compatible.

Fixed

  • MCP session creation now provisions the workspace and dataset when a fresh
    store reports them missing, then retries once. If catalog creation fails,
    callers receive the original session-start error rather than the create error.
  • Ending a session in default Mode 0 now processes deferred records without a
    consolidation loop, allowing finalization to complete.
  • Turkish legal rule comparisons and date lookups handle uppercase İ/I
    consistently with lowercase text. Original text, offsets and rule versions
    are preserved.
  • Scoped FTS queries include deduplicated ı → i variants, so YARGITAY
    and yargıtay find text written YARGITAY KARARI.
  • Context requests reject timestamps without timezone offsets with HTTP 422
    before comparing ranges. Fact validation treats date-only values as UTC
    when comparing them with timezone-aware timestamps, preventing datetime errors.
  • Gateway circuit breakers release half-open probes after permanent errors or
    cancellation. Non-retryable bridge writes enter a terminal failed state
    instead of being sent again on every drain.
  • Projection rebuild smoke checks use the same assertion text, including the
    evidence span, as vector replay.
  • Duplicate or stale projection completion calls preserve committed artifacts;
    artifact compensation remains restricted to fenced mutations.
  • SDK quickstart examples wait for ingestion before requesting context and close
    document files. Documentation links and the architecture version are corrected,
    and embedding setup notes explain downloading the model before enabling it.
  • Parser regression checks retain their 200 ms budget using thread CPU time,
    with a 2 s hang guard. Claude adapter tests skip when the optional anthropic
    package is absent.
  • Legal rule extraction skips impossible decision-reference scans and delays
    law-name cleanup until an attached article is found. On the API-limit
    TCK/1 input, best-of-five local thread CPU fell from 129.861 ms to
    52.330 ms (2.48x faster), with identical saved outputs and spans. The
    isolated 200 ms CI CPU-budget check is blocking again; the full local
    suite retains its 2 s hang guard.

Changed

  • Bulk ingestion batches SQL and graph projections and outbox completion in
    the existing worker pass. The measured 4,999,000-byte document reached full
    readiness in 245.20 s versus 718.27 s before (2.93x faster), with all 4,882
    mutations COMMITTED, SQL and GRAPH COMPLETED, VECTOR SKIPPED, and lossless
    chunk readback. Durability and fencing behavior are preserved.
  • The default embedding model is now local:BAAI/bge-m3, with 1024 dimensions
    and normalization enabled, including the Compose model default. Models remain
    off by default. Stores created with the old default must keep both
    MESA_EMBEDDING_MODEL=local:magibu/embeddingmagibu-200m and
    MESA_EMBEDDING_DIMENSION=768, or rebuild before switching embedding identity.
  • Tag-triggered PyPI releases wait for the complete reusable CI workflow before
    building and publishing.
  • Startup fails with a clear error when MESA_TIER3_MODE=1/2 has no active
    consolidation loop; extraction must be enabled for these modes.
  • Bootstrap MESA_API_KEY secrets containing . are rejected at startup.
    Placeholder API keys and principal IDs in .env.example are commented out.
  • Antigravity bridge write spools are isolated per workspace. Entries in the old
    shared spool are left in place and are not migrated automatically.

Known issues

  • Kuzu 0.11.3 can write uninitialized memory into graph page tails, so raw graph files may contain fragments of text the process handled; MESA never stores memory text in the graph (checked by query), and the graph directory should be treated as sensitive like the SQLite file.
  • Per-minute and daily rate limits are not enforced for /v4/* and /control/*
    in this version; the API documentation now states this limitation.
  • The ASCII query yargitay still does not match text written Yargıtay;
    index-side Turkish folding remains deferred.

Install and upgrade

pip install --upgrade mesa-memory==0.9.3

Follow the upgrade procedure when upgrading an existing store. Embedding identity migration requirements are listed above.

Validation

Full comparison.

Published artifacts

The attached wheel and source distribution match the files published on PyPI by SHA256. SHA256SUMS also covers the locked-runtime CycloneDX SBOM generated by the tag CI. A temporary install of the PyPI wheel passed API, SDK, storage, worker, and configuration imports plus mesa --help.

MESA v0.9.2 — legal retrieval and admission hardening

Choose a tag to compare

@Yasou13 Yasou13 released this 06 Nov 18:45
8cd2cfe

This release hardens the Turkish legal extraction and retrieval features from
Round 16 and Round 17. V4 HTTP routes remain compatible. Production qualification
remains NO-GO pending the normal certification and production-like release gates.

Added

  • Rule-based Turkish legal item extraction for newly admitted chunks, capturing
    versioned legal rule items with chunk offsets (citation, decision_ref,
    date, amount, duration, and outcome). Extracted items are returned in
    search provenance under legal_items separately from SPO facts, without
    affecting fusion scores. Document purge erases their source quotes.
  • Store-learned law names: authorized document titles and law headers provide
    scope-cached and indexed law names, including the K. abbreviation.
  • Exact-match-first slot: queries matching a decision's own docket/decision reference
    or exact law/article occupy at most three leading search positions in document order,
    preserving remaining result order. Decision matching requires references beginning in
    the first 300 characters of an authorized first chunk; later quotations do not count.

Changed

  • MESA_LEGAL_RULES_ENABLED defaults to true. When set to false, it disables
    automatic legal identity assignment and item extraction during ingestion. During
    search, it disables legal processing (citation resolution, exact-match slot,
    decision künye promotion, legal ranking factors, and graph citation parsing) and
    suppresses stored legal result overlays in returned copies while keeping durable
    identities and snapshots intact.
  • Open-ended SPO fact extraction (MESA_EXTRACTION_ENABLED) defaults to false
    regardless of MESA_MODEL_ENABLED; an explicit true still enables it.
  • Constitutional Court names (Anayasa Mahkemesi, its suffix forms and AYM)
    no longer activate Constitution citation ranking; separate statute citations in
    the same query are preserved.
  • Large-document admission waits for queue room on queue_over_capacity without
    repeating full document chunk planning.

Security

  • Fixed synchronous CPU-bound request-path stalls that also affect 0.9.1: rule
    extraction, citation resolution, scope building, and heavy token scoring now bound
    caller-text work and yield thread execution, keeping /health responsive under
    high load. Users of 0.9.1 are recommended to upgrade.

Known issues

  • Post-merge CI on Python 3.13: a 200 ms wall-clock parser test exceeded its bound on the CI runner (heaviest case ~162 ms locally). Outputs are unaffected; the speed-up lands on main and ships in 0.9.3.

  • Heavy 1,000-candidate search takes approximately 11.6 s; /health responsiveness
    is restored for measured repros, but heavy-query execution time is not proved faster.

  • Older whole-document secret scanning and extreme Unicode normalization can still
    cause sub-second request-path stalls.

  • UUID ties can change ordering across separately ingested stores; same-store
    retrieval remains deterministic for retained inputs.

  • Large documents take minutes to become fully ready: a 5 MB document
    (4,999,000 bytes) took about 7 minutes (7m00.6s) to become ready with models off
    and graph on.

  • Non-object document requests keep an INFO-level 422 detail wording difference;
    HTTP status, error code, location, and OpenAPI schema remain unchanged.

Install and upgrade

pip install --upgrade mesa-memory==0.9.2

Back up existing stores and follow the upgrade procedure before upgrading.

Validation

Full comparison.

The annotated v0.9.2 tag is SSH-signed and passed the full local signed-tag preflight.

Final PR coverage: Python 3.11 88.09%, Python 3.13 88.07%; 2,032 tests passed in each coverage job. Full local release gate: 2,037 local tests plus 10 separate MCP tests passed.

Build evidence

Attached distributions match the published PyPI files by SHA256. The build evidence archive contains the reproducible tag-CI wheel, locked-runtime CycloneDX SBOM, and verified OIDC provenance reports constrained to Yasou13/MESA, .github/workflows/ci.yml, refs/tags/v0.9.2, and release commit 8cd2cfe1df5356e73b3bef83ae640850fd54c798.

MESA v0.9.1 — faster retrieval and article-aware search

Choose a tag to compare

@Yasou13 Yasou13 released this 01 Nov 16:03

MESA v0.9.1 combines Round 13 (speed) and Round 14 (retrieval quality and document workflows), following v0.9.0. SQLite remains the source of truth; LanceDB and Kùzu remain derived projections. Existing V4 HTTP routes and request shapes remain compatible.

Round 13 — speed and operational fixes

  • Cache up to four authorized search scopes, including temporary BM25 indexes and derived text features. SQLite commit changes invalidate the cache; authorization is still checked on every request. MESA_SEARCH_SCOPE_CACHE=false disables reuse.
  • Add ingestion-ledger indexes and reconcile revision activation once per commit transition, keeping existing completeness, supersession and rollback rules.
  • Skip graph seed preparation when graph retrieval is disabled.
  • Move scrypt key verification, issuance and rotation off the event loop without caching verification results.
  • Make mesa add wait for an active revision by default; --no-wait returns after admission. Retry admission capacity errors within the SDK deadline.
  • Accept zero RRF lane weights, preserve timestamped runtime logging through migrations, and record projection start times from outbox claims.

Round 14 — retrieval quality and document workflows

  • Split Turkish legislation at article headings while preserving the exact source text and the 512-token chunk bound. MESA_CHUNKING=auto is the default; plain retains the previous plan for new revisions. Existing active revisions are unchanged.
  • Add optional lazy cross-encoder reranking with BAAI/bge-reranker-v2-m3, after fusion and outside the SQLite search connection. It remains off by default. Install the existing [ml] extra and set MESA_RERANK_ENABLED=true; default depth is 50. Results add rerank_score while retaining fusion scores.
  • Compute query embeddings before taking the shared search connection, allowing provider calls from concurrent searches to overlap.
  • Select default vector RRF weight 8.0 from the frozen Turkish Tier M dev split; other lane weights remain 1.0. Reranker-off passage MRR increased from 0.3953 to 0.5663 under the documented setup. These are corpus-specific dev measurements, not a general performance guarantee.
  • Report terminal chunk failures promptly with mutation ID, state and error class instead of waiting for the document timeout.
  • Delete documents by source with mesa delete --source <ref> or Mesa.delete(source=...); document-ID deletion remains supported.
  • Include OpenAI and Ollama SDKs in the base package. Anthropic remains in [adapters].
  • Add tag-triggered PyPI trusted publishing through GitHub Actions/OIDC.

Install and upgrade

pip install --upgrade https://github.com/Yasou13/MESA/releases/download/v0.9.1/mesa_memory-0.9.1-py3-none-any.whl
mesa start

Requires Python 3.11+. Linux and macOS are supported natively; use Docker or WSL on Windows. Models remain optional and off by default. The command above installs the validated wheel attached to this GitHub release. After PyPI publication is confirmed, the equivalent package-index command is pip install --upgrade mesa-memory==0.9.1. PyPI trusted-publisher setup is a separate owner configuration step.

Follow the backup and upgrade procedure before upgrading an existing store; normal startup migrations add the ingestion indexes. To retain earlier behavior for new revisions, set MESA_CHUNKING=plain; for the previous vector weighting, set MESA_RRF_VECTOR_WEIGHT=1.

An interrupted upload whose admitted prefix matches two different chunking plans fails closed with HTTP 409 rather than silently changing the plan. See configuration.

Evidence and status

  • Both implementation rounds passed local and GitHub checks before merge.
  • Release commit e2f984d4ad97c01782a347e29e44808e5a16dad3: full local gate PASS (1,860 tests and 10 optional MCP tests), 37 focused release tests, metadata preflight, wheel/sdist build and Twine validation PASS. MESA CI and external release gates both PASS.
  • Article coverage: 2,988/3,063 indexed starts (97.55%); all 141 Tier M files round-trip losslessly, with unchanged plain chunks for the 46 non-legislation files.
  • Full measured dev matrix, model/hardware details and limitations: retrieval benchmark.

This is a source release. Production qualification remains NO-GO until the normal certification and production-like gates are complete; CI success and dev retrieval results do not certify production readiness.

Tag-signing status: the owner explicitly approved an unsigned annotated tag for v0.9.1. Metadata preflight passed; signed-tag preflight success is not claimed. The tagged release commit is included in main via PR #141.

Full comparison: v0.9.0...v0.9.1

MESA v0.9.0

Choose a tag to compare

@Yasou13 Yasou13 released this 28 Oct 17:40
7062faf

MESA v0.9.0 brings together the MVP work delivered across Rounds 1–12: a smaller V4-only core, model-optional document search, durable processing, safer document lifecycle handling, and local CLI/SDK/MCP onboarding.

MESA stores source-linked evidence for AI applications. SQLite remains the source of truth; LanceDB and Kùzu are derived projections. MESA returns evidence with document, revision and passage provenance; the calling application writes the final answer.

Install and start

Requires Python 3.11+. Linux and macOS are supported natively; use Docker or WSL on Windows.

pip install "mesa-memory[adapters]==0.9.0"
mesa start

In another terminal:

mesa add document.txt --source file://document.txt
mesa search "your query"
mesa mcp-config

Models are off by default. Documents can be stored and found through BM25 without an LLM or embedding model. Optional models add fact extraction and semantic retrieval.

Highlights

  • Local single-user setup with generated credentials, loopback binding, and mesa start, add, search, delete, status, and mcp-config.
  • A convenient Python Mesa SDK and whole-document ingestion through POST /v4/documents.
  • Lossless chunking, deterministic chunk identifiers, duplicate detection, durable retry, and revision tracking.
  • Unified provider:model settings for LLMs and embeddings, plus local Ollama and OpenAI-compatible endpoints.
  • An optional graph backend, explicit degraded-source reporting, and backfill when disabled projections are enabled again.
  • A PyPI package, local Docker workflow, Colab quickstart, and rewritten onboarding documentation.

What changed in Rounds 1–12

Round Delivered changes
1 — Offline baseline and V4 contracts Bundled the cl100k_base tokenizer data so token counting works offline. Ported release-gate coverage for authorization, principal/session isolation, durable dispatch and finalization to V4. Removed obsolete V3 MCP tools and moved MCP health checks to /health; removed unused evaluation/script packaging and hooks.
2 — V3 and obsolete runtime cleanup Removed V3 HTTP routes and client methods, the split api-only/worker-only profiles and Compose topology, obsolete background/retrieval/tracing modules, and legacy ingestion/consolidation branches. Kept canonical V4 fact extraction and durable processing. Raised the supported Python floor to 3.11 while updating compatible security fixes in the lockfile.
3 — Dependency and configuration cleanup Removed unreferenced dependencies, obsolete extras, dead configuration fields and 13 unused MemoryDAO methods. Updated installation, CI and documentation contracts for the reduced dependency surface.
4 — Retry completion and queue location Fixed V4 records becoming stranded in RETRY_PENDING, terminal failure handling and false acknowledgments. Put runtime dead-letter files under the configured storage root.
5 — Recovery and extraction fallback Added spaced retry backoff, startup crash recovery and working dead-letter replay. Made LLM storage validation opt-in. Unusable extraction output falls back to storing the source chunk without facts; provider failures still surface as failures. Separated timing-sensitive tests from the normal local gate.
6 — Model-free storage and search Enabled writes and BM25 search without models, independent extraction/embedding switches, skipped-vector backfill and identity-safe recovery. Added degraded_sources to search responses and configurable Ollama timeout, context window and generation options.
7 — Unified model configuration Standardized MESA_LLM_MODEL and MESA_EMBEDDING_MODEL on provider:model, added Ollama embedding support and dimension detection, and preserved fail-closed embedding identity checks. Corrected documentation for model-optional writes and search.
8 — Document lifecycle and privacy Fixed rebuild after delete/rollback, rejected reused revision numbers and writes to purged documents, and kept deleted/replaced content out of session context. Tightened historical revision filtering. Purge removes document text while preserving identifiers, hashes and shared ownership. Provider credentials now come from MESA-specific variables. See the graph-file limitation below.
9 — MCP and graph hardening Enabled direct stdio MCP operation without an embedded control-plane database. Improved MCP errors and agent-facing responses, always wrapped recalled source text as untrusted evidence, and added a recommended answering-system prompt. Restart resumes interrupted dispatch/projection work. Graph assertion text stays canonical in SQLite, with graph ranking reading authorized SQLite evidence.
10 — Local first-use workflow Added the Mesa convenience SDK, local CLI and POST /v4/documents. Whole-document ingestion preserves exact source slices, revision identity and retry idempotency. Added MESA_GRAPH_ENABLED=false, graph-aware readiness and later graph backfill. Fixed CLI result rendering and recalled-evidence wrapping.
11 — Onboarding, local providers and packaging Improved Turkish abbreviation/list/date chunk boundaries; rejected attempts to restore superseded text with a clear 409 response. Added bounded SDK retries for transient startup 503s and actionable missing-extra messages. Centralized local-endpoint checks, supported keyless local model servers, and made Docker provision persistent local credentials. Prepared v0.9.0 packaging, Colab and rewritten README/docs, with fresh-install and Docker checks.
12 — PyPI release preparation Switched general installation examples to the published PyPI package, retained one development-from-Git example, added the Colab restart note, dated the changelog and completed the README installation step. Final round checks covered packaging, rendered metadata, existing test preservation and the full local/MCP gates.

Publication follow-up: corrected an invalid PyPI Trove classifier before the successful upload in PR #136. Both the wheel and source distribution are published on PyPI.

Compatibility and upgrade notes

  • V3 interfaces have been removed. Integrations should use the current V4 HTTP API, Python SDK or MCP tools. The retired split runtime profiles and removed configuration fields/extras should be removed from deployment configuration.
  • Use MESA_LLM_MODEL and MESA_EMBEDDING_MODEL for model selection, and MESA-specific credential variables for providers.
  • Existing stores need the documented offline projection rebuild to adopt the graph representation with assertion text fields empty. Review the rebuild runbook and configuration guide before upgrading.

Verification

  • Release commit: 7062faf.
  • Same-commit MESA CI and external release gates: passed. Tagged-release CI: passed.
  • Clean wheel/sdist build and Twine checks passed. PyPI file hashes match the locally verified release artifacts.
  • A fresh Python 3.11 environment installed mesa-memory[adapters]==0.9.0 directly from PyPI. The downloaded wheel hash, version, SDK imports, mesa --help and dependency consistency checks passed.
  • The v0.9.0 tag is annotated and unsigned.

MVP status and known limitations

This is the v0.9 MVP release. Production qualification remains NO-GO until the documented production release gates are satisfied; publication and CI success do not certify production readiness.

Kùzu rebuilds can leave fragments of live document text in unused file bytes, and those fragments may survive a later purge. Strict-erasure deployments can disable the graph with MESA_GRAPH_ENABLED=false and retire existing graph generations/backups according to their retention policy. See security and privacy for the precise limitation and rebuild guidance.

Getting started · MCP setup · Colab quickstart · Full changelog

MESA v0.8.1 — MVP hardening and developer experience

Choose a tag to compare

@Yasou13 Yasou13 released this 17 Oct 22:17

Publication status

The repository owner authorized this v0.8.1 source release on 2026-10-18.
It includes the MVP hardening and developer-experience changes merged in PR #112.
Production qualification remains NO-GO pending the normal release gates;
real LLM/embedding integration and a production-like 24-hour soak are not certified.
No local GPG signing key is available, so the annotated release tag is unsigned.
Signed-tag preflight success is not claimed. CI, wheel/SBOM artifacts and OIDC
attestations are verified separately for the release commit and tag.

Removed

  • Removed the obsolete standalone raw-log agent-ID backfill script.
  • Removed obsolete manual backup/restore, MCP spoofing, Mem0 smoke and async I/O
    benchmark scripts from tests/; removed their stale collection exclusions.
  • Removed the unused dashboard/showcase frontend, its npm dependencies and build
    configuration, and the obsolete run_demo_rag.py launcher. MCP control APIs
    and operator approval CLI remain available; see control API.

Fixed

  • Scoped BM25 corpus statistics, pre-rank audit queries and audit metadata to
    authorized tenant/dataset scope; unrelated data no longer influences results.
  • Protected projection completion with claim/fencing ownership checks; prevented
    cross-dataset repair deletions and false invalidation when vector lookups fail.
  • Kept Kùzu caches consistent after delete and transaction rollback, including
    composite assertion IDs; graph commit failures retry and cancellations roll back.
  • Surfaced vector backend and embedding failures explicitly; removed the legacy
    raw-memory fallback to an unrelated eight-dimensional embedding.
  • Rejected reserved _mesa_ metadata and non-finite RRF weights while preserving
    explicit zero confidence. Mapped context, catalog chunk and rollback errors
    to actionable API responses instead of generic server errors.
  • Added optional legal_identity parity to synchronous and asynchronous V4 SDK
    insertion; slowed default status polling to fit the shared minute request budget.
  • Returned field-specific validation errors without echoing rejected input or
    internal route paths, and retained startup error logging after migrations.
  • Corrected standalone runtime defaults and documented a minimal supported
    Python install. Updated CI import checks, documentation contracts and SQLite
    fault-injection test doubles.
  • make health now checks the running API's readiness and authenticated health
    using MESA_BASE_URL and MESA_API_KEY, without opening local databases.
    Unhealthy responses and connection/authentication failures return a nonzero exit.

Documentation

  • Clarified provider prerequisites, immutable embedding identity, asynchronous
    projection status, source evidence, session finalization and client cleanup.
    The V4 quickstart displays assertion/source references. Models-disabled startup
    is described as a health check; V3 ingestion still requires embeddings.

  • Verified small SQLite/LanceDB/Kùzu flows, restart, populated tenant isolation
    and completed rollback with explicitly controlled model boundaries. These
    checks do not establish real-model quality or production-size capacity.

  • Consolidated 37 Markdown pages into 20 maintained guides and decision records;
    merged overlapping guides, retired obsolete ADR files and repaired document
    contracts and links. Release-specific exceptions remain in this changelog.

  • Consolidated documentation ownership and navigation; replaced the root V3
    whitepaper with a current architecture entry point.

  • Removed obsolete Colab guidance, the retired architecture/audit/benchmark
    documents and their archive indexes. Merged the external benchmark ownership
    note into the ecosystem guide and repaired incoming documentation links.

  • Removed duplicated setup/recovery guidance, corrected MCP authorization and
    rebuild/migration wording, and clarified ADR applicability. These documentation
    changes do not certify a new release; prior tracked records remain in Git history.

Release verification

  • Release commit: d394784.
  • Local: 108 focused tests, repository-wide Ruff, scoped CI Black, Mypy across 138 production files, lock/metadata/documentation checks and an isolated installed-wheel smoke passed.
  • Passed same-commit workflows: MESA CI and external release gates.
  • Tag package/provenance job succeeded; downloaded artifact hashes and provenance subject digests were matched to the release commit.
  • Assets contain the CI-built 0.8.1 wheel, runtime CycloneDX SBOM, GitHub OIDC provenance bundles and SHA256 checksums.
  • Real-provider integration, production-size capacity and a 24-hour soak remain outside this source release.

Full changelog: v0.8.0...v0.8.1

MESA Core v0.8.0

Choose a tag to compare

@Yasou13 Yasou13 released this 15 Oct 18:21
07d387f

MESA Core v0.8.0 release notes

Release date: 2026-10-15. Package version: 0.8.0.

The repository owner explicitly approved publication after reviewing the missing
certification evidence. This source release does not certify production readiness.
Actual publication and artifact status are recorded in GitHub Releases and the
release audit; a version bump alone is not publication proof.

Verification and known limitations

  • The documentation PR108 and version PR109 are checked independently by MESA CI
    and external release gates; exact commit/run results are linked below.
  • Real model/provider boot/restart/token-cost evidence, a production-like 24-hour
    soak and new official Profile B certification are not supplied by this release.
  • Full per-method docstring certification and the pre-existing full-tree Black
    formatting debt remain incomplete; required scoped CI checks are preserved.
  • No local GPG signing key is available: the annotated v0.8.0 tag is unsigned.
    Signed-tag preflight success is not claimed. Build/SBOM/OIDC provenance status
    is verified separately and must not be inferred from publication alone.
  • Production status remains NO-GO until the normal certification gates pass.

Changes

The development history since v0.7.1 contains selectable validation, canonical
embedding identity, durable offline projection rebuild, immutable revision
activation/ownership recovery, four-lane scoped retrieval, graph runtime closure,
atomic evidence packing, bounded opt-in query planning and typed legal identity.
The current documentation refresh aligns entry points, API examples, provider
defaults, configuration, ecosystem boundaries and limitations; removes unsourced
showcase metrics/PASS output; and adds scoped neutral V3 and provider-required V4 examples.

Independent review corrected conservative offline tokenizer counting, low-level
vector versus canonical authorization, V3 router descriptions and smoke checks
under Python optimization. Dashboard build dependency lock patches address two
detected HIGH advisories; audit/build proof remains separate from production gates.

See CHANGELOG v0.8.0 for the detailed implementation
history, migration for forward schema/representation
requirements, and release policy for immutable signed tagging,
same-commit CI, provider, backup/rebuild and 24-hour soak requirements.

No new model accuracy, legal correctness, official Profile B PASS, native
MESA_Data delivery, production-size memory bound or 24-hour soak result is
claimed. Publication proof and exact test results belong to the release audit run
under reports/v0.8.0-documentation-release/; an absent tag/release is not success.

Verification references

No PyPI/npm/container registry publication or production deployment is part of
this GitHub release. See the attached checksums and provenance for exact artifact
identity; these do not certify model accuracy or 24-hour stability.

Release artifact provenance

The wheel and CycloneDX 1.6 runtime SBOM were built by the tag CI workflow.
Their Sigstore/OIDC attestations were cryptographically verified against
Yasou13/MESA/.github/workflows/ci.yml, refs/tags/v0.8.0 and source commit
07d387f9a4592ea39518fea549e9cf8330051e1f, with self-hosted runners rejected.
See the attached SHA256SUMS and Sigstore bundles. This artifact provenance is
separate from the unsigned Git tag and the missing production certification.

gh attestation verify mesa_memory-0.8.0-py3-none-any.whl \
  --repo Yasou13/MESA --source-ref refs/tags/v0.8.0 \
  --source-digest 07d387f9a4592ea39518fea549e9cf8330051e1f

Exact release-commit workflow runs:

Release commit: 07d387f9a4592ea39518fea549e9cf8330051e1f. The annotated tag is unsigned under the explicit owner publication exception.

MESA v0.7.1

Choose a tag to compare

@Yasou13 Yasou13 released this 29 Jul 08:36

MESA v0.7.1

MESA v0.7.1 is a maintenance patch release that packages the latest production hardening and CI security fixes on top of v0.7.0.

Highlights

  • Updated package, Docker, SDK, evaluation, benchmark, and documentation version references to v0.7.1.
  • Aligned active source-code, test-suite, and benchmark labels with the current release version.
  • Includes the merged production hardening and CI security improvements from the main branch.
  • Preserves historical v0.6.1 documentation and compatibility references where they describe prior releases or dependency constraints.

Upgrade Notes

This release does not introduce new public APIs or database migrations. Existing v0.7.0 deployments can upgrade directly to v0.7.1.

Verification

The release metadata and active version references were validated with dependency-lock checks and the affected test suite.

MESA v0.7.0

Choose a tag to compare

@Yasou13 Yasou13 released this 23 Jul 19:47

MESA v0.7.0 — v4 Full-Cognitive Release Candidate

MESA v0.7.0 introduces the new v4 full-cognitive runtime while preserving the existing v3 lexical-core API for compatibility.

Highlights

  • New versioned v4 API, Python SDK, and MCP contracts
  • Durable mutation ledger and projection outbox
  • Ordered SQLite → vector → graph projection pipeline
  • Graph V2 with canonical entities and provenance-rich assertions
  • Tenant, workspace, and dataset-level authorization
  • Source-aware artifact ownership, rollback, purge, and reconciliation
  • True Reciprocal Rank Fusion across vector, BM25, and graph retrieval
  • Single-writer combined runtime for SQLite, LanceDB, and Kùzu
  • Rotation-safe, principal-bound API keys using salted scrypt hashes
  • Pipeline leases, heartbeat renewal, bounded retries, replayable DLQ, and crash recovery

Canonical Data and Graph V2

Entities now use deterministic, tenant-scoped UUID5 identities derived from ontology identifiers or normalized canonical names.

Graph relationships are represented as immutable assertions carrying:

  • Predicate and subject/object identity
  • Source document, revision, and chunk
  • Evidence span and citation
  • Jurisdiction and source authority
  • Confidence and temporal validity
  • Mutation and pipeline-run provenance
  • Contradiction and supersession relationships

Different predicates, sources, jurisdictions, and time ranges remain independently queryable.

Durable Ingestion

POST /v4/memory/insert creates a durable mutation and pipeline run before background processing begins.

Mutations progress through an idempotent lifecycle:

RECEIVED → EXTRACTED → VALIDATED → SQL_APPLIED → VECTOR_APPLIED → GRAPH_APPLIED → COMMITTED

Tier-3 validation is now a mandatory gate. Rejected information does not create active SQL, vector, entity, edge, or assertion artifacts.

Artifact Ownership and Rollback

Physical artifacts can be shared by multiple mutations and sources. Rolling back or purging one source removes only its ownership link.

Shared entities and vectors remain active while another valid source still owns them. Artifacts are physically removed only after their final active owner is removed.

Rollback and cleanup operations are durable, restart-safe, and tracked through the outbox.

Security

V4 authorization is resolved server-side through:

Principal → Tenant → Workspace → Dataset → Agent → Session

Agents remain computational or persona context and are not treated as tenant security boundaries.

The release adds:

  • Tenant, workspace, and dataset memberships
  • OWNER, WRITER, and READER roles
  • Explicit purge and rollback permissions
  • Server-bound session scopes
  • Cross-tenant and cross-dataset isolation
  • Multiple keys per principal
  • API-key rotation and revocation
  • Salted scrypt hashes with key-ID lookup

Plaintext API keys are available only when initially created.

Retrieval V2

V4 retrieval combines vector, BM25, and graph rankings using true rank-based Reciprocal Rank Fusion.

Results can be filtered by authorized datasets and reranked using bounded legal and provenance signals, including:

  • Jurisdiction
  • Effective time
  • Source authority
  • Citation availability
  • Assertion confidence

Deterministic tie-breaking is preserved.

Runtime and Operations

V4 supports one storage-writing combined process that owns SQLite, LanceDB, and Kùzu. Multiple processes must not write to the same storage directory.

Operational improvements include:

  • Fenced claims and lease renewal
  • Bounded retry policies
  • Replayable dead-letter handling
  • Ordered projection prerequisites
  • Two-way reconciliation
  • Shared-owner and orphan metrics
  • Rollback and cleanup backlog visibility
  • Offline backup, rebuild, parity, and cutover tooling

PageRank is now observation-only telemetry and cannot quarantine or invalidate memory artifacts.

Compatibility

  • V4 is a breaking, versioned API.
  • Existing v3 endpoints remain unchanged.
  • V3 continues as the lexical-core compatibility runtime.
  • V3 storage is not migrated in place.
  • Legacy data must be backed up and rebuilt into separate v4 storage before cutover.

Release Status

MESA v0.7.0 is a release candidate and remains NO-GO for production until all external release gates are completed, including real-provider validation, production-like crash and concurrency evidence, migration and restore rehearsal, and the required 24-hour soak test.

MESA v0.6.1

Choose a tag to compare

@Yasou13 Yasou13 released this 17 Jul 13:11

MESA v0.6.1 Release Notes

This release establishes a stable version lock prior to impending architectural changes in the repository. The update strictly focuses on test coverage expansion to meet CI thresholds, type safety enforcement, and the finalization of remediation phases.

Features & Remediation

  • Completed Phase 1 to 5 remediation implementation.

Type Safety & Bug Fixes

  • MCP Server: Resolved mypy type check errors to ensure static analysis compliance.
  • Core: Corrected mypy type hint for the mget return array.

Testing & CI Pipeline

  • Coverage Compliance: Added comprehensive test coverage for DAO and API router error paths, successfully pushing project coverage past the 85% threshold.
  • Test Refactoring: Rewrote benchmark and ingestion tests with corrected class and function names.
  • CI Stability: Restored previously missing coverage tests to satisfy CI pipeline constraints.

Documentation & Maintenance

  • Documentation: Synchronized architecture documentation, changelog, and README to reflect v0.6.0 features accurately.
  • Code Quality: Applied black formatting to test_dao_coverage.py.
  • Cleanup: Removed redundant files and cleaned up scripts within the root directory.

---This is a release created to maintain the repository.

MESA v0.6.0

Choose a tag to compare

@Yasou13 Yasou13 released this 17 Jul 07:19

MESA v0.6.0: Comprehensive Audit Remediation

This release addresses 49 findings identified during the MESA architectural and security audit. It introduces critical security patches, data integrity enforcements across the storage layers, and the final deprecation of legacy interfaces.

⚠️ Breaking Changes:

  • API Authentication: Client integrations must now use the X-API-Key header standard instead of Bearer tokens.
  • Storage Layer: All legacy StorageFacade endpoints and methods have been permanently removed and unified under MemoryDAO.
  • Search Schema: SearchResultItem payloads now mandate the inclusion of content_payload.

🔴 Critical Security & Integrity Fixes

  • [R-01] IDOR Prevention: Added strict agent_id validation to soft_delete and hard_delete methods in vector_engine.py to prevent unauthorized vector manipulation.
  • [R-02] ASCII Date Comparison: Fixed SQLite DELETE queries during soft-delete purging to strictly cast and compare valid datetime objects.
  • [R-03] Dual-Write Saga Safety: Implemented _atomic_saga_commit in dao.py. SQLite commits are now strictly deferred until asynchronous Vector and Graph engine writes succeed, rolling back upon failure to prevent split-brain anomalies.
  • [R-04 & R-11] Endpoint Security: Applied Depends(get_api_key) decorators to /v3/health and /v3/demo/chat to close unauthenticated DoS and unauthorized access vectors.
  • [R-05] XSS/Data Sanitization: Enforced _sanitize_payload universally across all DAO read operations, filtering content and content_payload through sanitize_cmb_content before returning data.
  • [R-07] KùzuDB Connection Leaks: Enclosed KùzuDB execute operations within try/finally blocks, ensuring QueryResult objects are deterministically closed for read-only queries to prevent lock exhaustion.
  • [R-10] Prompt Injection Guard: Removed agent_id from MCP tool arguments. The system now strictly reads MESA_AGENT_ID from environment variables, preventing LLM impersonation attacks.

🟠 High Priority Performance & Reliability

  • [R-12] Graceful Shutdown Leaks: Added cancel_futures=True to the ThreadPoolExecutor.shutdown call in vector_engine.py to prevent thread hangs during teardown.
  • [R-13] CrossEncoder OOM Prevention: Bound the reranker.py thread pool strictly to max_workers=1, preventing Out-Of-Memory exceptions during high-concurrency retrieval spikes.
  • [R-15] N+1 Ingestion Bottleneck: Refactored REBEL triplet ingestion to utilize compute_embedding_batch prior to loops, eliminating N+1 embedding call overhead.
  • [R-17] Observability: Integrated dynamic LangFuse/LangSmith telemetry tracing via mesa_memory/observability/tracer.py attached to the FastAPI lifespan.
  • [R-18] Non-Idempotent Retry Safety: Disabled automatic retry mechanisms for non-idempotent HTTP methods (e.g., POST) in the MESA client SDK to prevent data duplication on timeout.

🟡 Medium Priority Logic & Workflow

  • [R-20] Cold-Start Hydration: Fixed ValenceMotor initialization by injecting storage=self.dao, allowing successful retrieval of pre-existing embeddings from LanceDB during startup.
  • [R-21] Utility & Novelty Filtering: Overhauled the compute_memory_valence algorithm to evaluate composite fitness (novelty + utility). Data scoring below 0.3 is now decisively discarded to prevent vector bloat.
  • [R-23] DLQ Reprocessing: Implemented a background worker for Dead Letter Queue (DLQ) records to automatically retry failed asynchronous consolidation tasks.
  • [R-25] Hybrid Search Latency: Transitioned FTS5 lexical queries to execute fully asynchronously in parallel with Vector and Graph retrieval via asyncio.gather.
  • [R-26] LLM Judge Determinism: Introduced ensemble_size=3 and majority voting to llm_judge.py to enforce self-consistency in benchmark evaluations.
  • [R-27] Legal Mode CI/CD Check: Added explicit cost/latency telemetry for legal_domain_mode and integrated legal_audit.py into the gatekeeper.py CI/CD pipeline to detect graph poisoning.

🟢 Maintenance, Deprecation & Docs

  • [R-30] Flaky Tests Removed: Stripped non-deterministic asyncio.sleep calls from tests, replacing them with proper event loop yielding (sleep(0)) or direct mock assertions.
  • [R-31] Secure Installer: Refactored install.sh to download fully to /tmp/ before execution, removing the unsafe curl | sh pattern.
  • [R-34] Deep Health Validation: Replaced the superficial readiness flag in /health/init with deep driver-level validations, including a live MATCH (n) test against the graph engine.
  • [R-33 & R-43 to R-49] Legacy Code Purge: Completely removed all residual StorageFacade logic, unused schema functions, and dead code artifacts from the codebase and test suite.
  • [R-35] Dynamic Maintenance: Transitioned VACUUM scheduling to rely on the MESA_VACUUM_HOURS environment variable for better DevOps control.