Skip to main content
Glama
Review MCP Server

Review MCP Server

by ottonic

Ottorev — Review MCP Server

The memory an engineering assistant needs. This local stdio MCP server records names, notes, tags, and decisions against an exact CAD revision in SQLite, so an assistant can pick up where it left off — remember that a face is the mating surface, that a tolerance callout is tight, or that a review concluded hold with the requirements behind it.

It never opens the CAD file: it stores the revision the caller supplies and compares it on recall, so it works on a laptop with no CAD software and no other server running.

Quick start

Requires Node.js 26+ and npm. From this repository:

npm install
npm run build
REVIEW_MCP_DB_PATH=/absolute/path/to/review.db npm start

For any MCP client:

{
  "mcpServers": {
    "review": {
      "command": "node",
      "args": ["/absolute/path/to/review-mcp-server/dist/src/index.js"],
      "env": { "REVIEW_MCP_DB_PATH": "/absolute/path/to/review.db" }
    }
  }
}

REVIEW_MCP_DB_PATH defaults to ./review-mcp.db; parent directories are created automatically. Records persist across restarts — protect and back up the database as application data.

Related MCP server: engineering-mcp

Tools

Tool

Behavior

remember

Save a named feature (re-remember replaces) or an anonymous note (re-remember appends) against an entity set, with tags. Returns replaced so a caller can tell which happened.

recall

Retrieve by artifact, entity, free text, or tags; marks each record stale against a supplied current revision. Follow pagination for complete retrieval.

remember_decision

Save the disposition reached about a revision and its basis — one row per requirement-check, naming the source, the evidence, the state, and the engineer's verdict. Re-remembering the same subject replaces it.

recall_decisions

Return decisions newest-first, each marked stale when it belongs to another revision. An entity_id filter matches only the supplied revision.

forget

Delete a named feature, an anonymous note by entity, or a decision by subject. Idempotent.

Results provide structured content and text; tool failures set MCP isError. Writes change local records; the client controls when to request those mutations. The server also advertises engineering-review prompts that compose with a geometry server, while the core record operations run on their own.

Artifact and subject semantics

An input artifact contains revision (sha256: plus 64 lowercase hex digits), format: "step", and optional URI and units:

{
  "revision": "sha256:0123456789abcdef0123456789abcdef0123456789abcdef0123456789abcdef",
  "uri": "file:///work/part.step",
  "format": "step",
  "units": "mm"
}

Supply the actual revision from the producer; the example digest is illustrative. A URI groups records by locator; without a URI, the revision is the lookup key.

uri accepts either a URI or a plain filesystem path, because a path is what a caller usually has in hand:

{ "uri": "/work/part.step" }

Both spellings address the same records: a path is normalised to a file: URI before anything is stored or matched. A relative path resolves against the server working directory. Normalisation never opens the file, so a record can be written for an artifact this server cannot read.

An entity filter is revision-scoped; a text or tag filter is not. An entity ID denotes different geometry in a different revision, so recall and recall_decisions match an entity_id only against records written for the supplied revision, and match it exactly — face:3 does not match face:30. Names, notes, subjects, and tags are matched across revisions, which is how an earlier record is found at all, and every result says whether it is stale.

Each returned record contains one artifact reference, its entity IDs, and the ID scheme. The entity IDs are opaque outside that scheme and revision.

Decision records

A review names or annotates geometry. A decision is the other thing a review produces: a disposition about a revision, and the basis it rests on.

One basis item is one requirement-check, and it answers for itself:

field

what it carries

requirement

what has to hold, stated so it can be checked

source

where the requirement came from: model-pmi, drawing, shop-spec, supplier-cert, standard, prior-decision, engineer-stated, or assistant-proposed

sourceDetail

the clause, note number, or document a reader can follow

evidence

the value as it was reported, quoted with its provenance

state

satisfied, violated, unresolved, conflicting, or not-applicable

question

required for unresolved and conflicting: what would settle it

faces

the geometry the item is about, in this revision; always present, empty when the item is about the part

verdict

the engineer's accepted / rejected judgement, with a reason when they gave one

unresolved and conflicting are first-class outcomes: a decision that carries them is a complete record of where the work stands, not a failure to decide. A basis item is refused when it cannot answer for itself — no source, or a state of unresolved/conflicting with no question.

Three properties matter more than the shape:

  • A decision is bound to a revision. The revision is part of its key, so the same subject at the same revision is recalled rather than reached again, and a decision made against a different revision comes back marked stale: true, still carrying the revision it belongs to.

  • Face IDs do not travel between revisions. Topological naming is unsolved in OCCT, and this API does not pretend otherwise: an entity_id filter matches only the revision that produced the ID.

  • Verdicts are what make a requirement agreed. Recording a verdict turns a line the assistant proposed into one the engineer accepted or rejected — the record of how much authority the assistant actually earned.

A stored decision is the caller's own account: the requirement as stated and the evidence as quoted. This server never measures anything, so a recalled decision is a record of what was concluded, not a measurement cache — Geometry remains the authority for numbers.

Limits

Review stores authored context: names, notes, and tags bound to an artifact revision, and decisions with the requirements they rest on. It does not store verified measurements, a complete immutable audit history, or authoritative approval records, and it does not compute geometry, run simulation, or decide manufacturability. Staleness is judged from a revision the caller supplies; Review never opens the CAD file.

There are no schema migrations: the database is a working file, and a changed table shape means deleting it. The server reports the version mismatch rather than guessing at a transformation.

Development

Requires Node.js 26+ and just. A Nix flake and .envrc provide the toolchain; run direnv allow after changing either.

Recipe

Purpose

just setup

Install dependencies.

just dev

Run TypeScript over stdio.

just build

Compile into dist/.

just test

Run unit tests.

just lint

Run oxlint.

just fmt

Rewrite formatting with Prettier.

just fmt-check

Verify formatting.

just typecheck

Run tsc --noEmit.

just check

fmt-check + lint + typecheck + test.

just clean

Remove compiled output, retaining the database.

Server logs go to stderr as newline-delimited pino JSON. MCP_LOG_LEVEL (debug|info|warn|error), MCP_LOG_FORMAT=human, and MCP_TRACE_ID tune them; secret-like fields are redacted.

Packaging

npm pack builds first and includes compiled runtime code, package metadata, and this README. Tests and source maps are excluded. Install the tarball to check the review-mcp executable with a compatible MCP client. The server serves both protocol eras: the 2025-11-25 handshake that stock MCP hosts negotiate, and the 2026-07-28 envelope through server/discover.

License

MIT. Third-party notices: THIRD_PARTY_NOTICES.md.

Available Tools

5 tools
forgetForget a ReviewB
Destructive

Delete a saved record. Pass the feature name to remove a named feature, an entity id to remove the anonymous notes attached to that entity, or a subject to remove a decision. Idempotent. Use only when the user explicitly asks to remove it.

Inputs: artifact (required), name, subject, entity_id. Returns: name, deleted, subject, entity_id

recallRecall ReviewsA
Read-onlyIdempotent

Retrieve saved reviews for a STEP file, optionally filtered by an entity, free-text search, or tags. Pass the current ArtifactRef from Geometry inspect; its revision is used to mark each returned record stale or current. Each record returns one artifact reference, its entity IDs and the ID scheme; use these together when handing an annotation to another tool. An entity filter matches only reviews written against the supplied revision, because a face ID denotes different geometry in a different revision; text and tag filters are not scoped that way. Use early in a conversation to restore persistent context about a part. Uses the supplied artifact revision to mark saved records stale; it does not open the artifact or require shared filesystem access.

Inputs: artifact (required), tags, limit, offset, +2 more. Returns: reviews, current_revision

recall_decisionsRecall DecisionsA
Read-onlyIdempotent

Retrieve decisions saved for an artifact, newest first, each marked stale when it was made against a different revision than the one supplied. Pass the current ArtifactRef from Geometry inspect. Filter by subject text, status, or an entity that appears in the basis. An entity filter matches only decisions made against the supplied revision, because a face ID is scoped to the revision that produced it: a decision about face:3 in an older revision is not a decision about face:3 now. Use this when a revision reappears, to recover what was already concluded instead of reaching it again.

Inputs: artifact (required), limit, offset, status, +2 more. Returns: decisions, current_revision

rememberRemember a ReviewA

Save a durable review about geometry: a semantic feature (give it a name, e.g. 'mounting holes') or a plain note (no name). Re-remembering a named feature REPLACES it (use this to correct the AI). Re-remembering the same entities without a name APPENDS facts. The more you save, the more the system understands the part across sessions.

Inputs: artifact (required), entities (required), entity_id_scheme (required), name, +2 more. Returns: review, replaced

remember_decisionRemember a DecisionA

Save the disposition reached about an artifact revision, with the basis it rests on: a list of requirement-checks, each saying where its requirement came from and what happened when it was checked. Re-remembering the same subject on the same artifact revision REPLACES that decision, which is how a decision is revised as the engineer answers questions or disagrees with a line. A different subject starts a separate decision, because one part can carry several at once. Record each engineer verdict when they have given one: a requirement the engineer accepted is a different thing from one that was proposed. Store each requirement as stated and each piece of evidence as it was reported. This server never opens or measures the artifact, so recall returns what was concluded, not a measurement — Geometry remains the authority for numbers.

Inputs: basis (required), subject (required), artifact (required), confidence (required), +3 more. Returns: decision, replaced

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.9/5.0

    Scored across 5 tools

    Disambiguation4/5

    The set is organized as two parallel pairs: recall/remember operate on reviews (features and notes), while recall_decisions/remember_decision operate on decisions with requirement-checks as a basis. Descriptions make the boundary explicit, though the near-identical verb stems mean an agent skimming names could still pick recall when it wants recall_decisions. forget is the only tool spanning both domains, but its description clarifies how to target features, entity notes, or decision subjects.

    Naming Consistency4/5

    Names are all lowercase snake_case with verb-first construction, and the pattern is readable: base verbs (recall, remember, forget) for reviews, and the same verbs plus a _decision(s) suffix for decisions. Minor deviation: singular remember_decision paired with plural recall_decisions breaks the symmetry, and forget has no _decision counterpart despite handling decisions.

    Tool Count5/5

    Five tools is tightly scoped to a persistent-memory server: two write paths, two read paths, and one delete. Each tool earns its place and there is no redundancy. Nothing feels missing or bloated for the stated purpose.

    Completeness4/5

    Full lifecycle is covered for both domains — create (remember, remember_decision), read (recall, recall_decisions), delete (forget) — with 'update' handled through documented replace-on-rewrite semantics, which is a valid design. Minor gaps: there is no way to enumerate or export all saved reviews/decisions without supplying an ArtifactRef, and no cross-artifact or cross-revision listing operation.

    Maintenance

    ActivityMaintained
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Enables multi-project workspaces to share structured notes, API contracts, and handoff messages via a local SQLite database, with versioning and read tracking.
      GPL 3.0
    • A
      license
      B
      quality
      D
      maintenance
      Enables local engineering workflow management by consolidating tickets, QA evidence, time tracking, root cause investigation, knowledge, and reporting into a single SQLite database, allowing generation of complete ticket packages for handoffs, dailies, or career evidence.
      23
      14 npm
      MIT
    • A
      license
      B
      quality
      C
      maintenance
      Enables security analysts to carry structured investigation state across sessions by storing typed observations, hypotheses, tasks, evidence and versioned decisions in a local SQLite graph with full-text search. It also enforces authorized scope and request controls, redacted and content-hashed evidence, duplicate-test signatures and validation gates that leave findings in draft until reproducibly confirmed, then produces technical and executive reports for Antigravity, Codex and other stdio MCP clients.
      20
      1
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables clients to index and search reviewed Markdown and text documents with SQLite FTS5, read sources and prompts, manage explicit notes, and query a local Ollama assistant that uses permitted tools.
      4
      MIT