# tc39-mcp (npm · tc39-mcp)

Independent MCP server for the TC39 specs (ECMA-262 + ECMA-402): clauses, search, diffs, history.

- Trust score: 81/100 (high trust)
- Change this week: +28
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-03

## Components

- remote · `mcp.xyzzylabs.ai`: 63/100, [markdown](https://verifymcp.io/servers/xyzzylabs-tc39-mcp/tc39-mcp.md), [page](https://verifymcp.io/servers/xyzzylabs-tc39-mcp/tc39-mcp)
- npm · `tc39-mcp`: 81/100 (this document), [markdown](https://verifymcp.io/servers/xyzzylabs-tc39-mcp/tc39-mcp-2.md), [page](https://verifymcp.io/servers/xyzzylabs-tc39-mcp/tc39-mcp-2)

## Channel facts

- Registry: `npm`
- Package: `tc39-mcp`
- Version: `0.6.1`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-08-03.

- **Supply Chain Security**: 86/100
  - No malware found by supply-chain analysis.
  - Only part of the dependency tree could be resolved (94 of 98), so this covers what we could see, not the whole tree.
  - No install/post-install scripts declared.
  - Only part of the dependency tree could be resolved (94 of 98), so this covers what we could see, not the whole tree.
- **Provenance & Transparency**: 97/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to xyzzylabs/tc39-mcp).
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 26 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 85/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (excellent).
  - Tool/resource definitions use about 7163 tokens (~51/item across 139 items; 19 tools + 120 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 23/100
  - Stability observed for 7 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 99% of tool parameters carry a description.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add xyzzylabs-tc39-mcp -- npx -y tc39-mcp
```

### Codex

```bash
codex mcp add xyzzylabs-tc39-mcp -- npx -y tc39-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "xyzzylabs-tc39-mcp": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "tc39-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add xyzzylabs-tc39-mcp --command npx --arg -y --arg tc39-mcp
```

### Hermes

```yaml
mcp_servers:
  xyzzylabs-tc39-mcp:
    command: "npx"
    args: ["-y", "tc39-mcp"]
```

### Other

```json
{
  "mcpServers": {
    "xyzzylabs-tc39-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "tc39-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-08-03 (score 81, +1)

No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-08-02 (score 80, +49)

- [security improvement] Install scripts: unverified → pass
- [security improvement] Provenance: unverified → pass
- [security improvement] Known CVEs: unverified → partial
- [security improvement] Malware scan: unverified → pass
- [security] Stability: Stability not yet verified: we do not have a sandbox capture of the MCP schema this version of the package serves yet.
- [security] The attested source repository moved: xyzzylabs/tc39-mcp
- [functional regression] Capabilities: pass → unverified
- [functional improvement] Stability: unverified → 0.20
- [functional improvement] Schema quality: unverified → excellent
- [functional improvement] License: unverified → pass
- [functional improvement] Dependency health: unverified → partial
- [functional improvement] Maintenance: unverified → pass
- [functional] Licence: MIT

### 2026-08-01 (score 31, +26)

- [security] Stability: Stability not yet verified: not enough scan history yet (needs a 30-day window).
- [functional improvement] Schema quality: unverified → 100
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Tool coverage: unverified → 100

### 2026-07-31 (score 5, −30)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-30 (score 35, −18)

- [security regression] Malware scan: pass → unverified

### 2026-07-27 (score 53)

First indexed and scored.

## MCP tools (19)

### `spec.about` (~71 tokens)

About this server

Return self-description of this MCP server: package name + version, per-snapshot pin metadata (sha, fetched_at, biblio_commit, clause_count) for every supported (spec, edition), plus test262 + proposals index headers when present. Lets callers verify freshness and reproducibility without loading the parses themselves.

### `spec.snapshots` (~109 tokens)

List spec snapshots

List every (spec, edition, sha, fetched_at) snapshot this server has parsed. Use to discover what historical SHAs you can query via `at: "<sha>"`, or to verify reproducibility across server versions. Optional `spec` / `edition` filters.

Input parameters:

- `edition` (string): Filter to one edition (concrete name like 'main' or 'es2025'). Omit for all.
- `spec` (string): Filter to one spec. Omit for both.

### `clause.get` (~195 tokens)

Get spec clause

Fetch a parsed TC39 clause as structured JSON: metadata, signature, algorithm steps, notes, cross-refs, and outward citations to external specs (Unicode, IETF, WHATWG). `spec` selects '262' (default) or '402'. `edition` defaults to `latest` (current stable release on both specs — es2026 today).

Input parameters:

- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `id` (string, required): Spec clause id, e.g. 'sec-tonumber' (262) or 'sec-intl.numberformat' (402).
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `clause.list` (~265 tokens)

List spec clauses

List parsed spec clauses with optional filters (kind, section prefix, has_algorithm). Returns lightweight rows {id, aoid, title, number, kind, algorithms}; follow up with clause.get for detail. `spec` selects '262' or '402'.

Input parameters:

- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `has_algorithm` (boolean): If true, return only clauses with at least one `<emu-alg>`.
- `kind` (string): Filter by clause kind (e.g. 'op', 'sdo', 'built-in function', 'concrete method').
- `limit` (integer): Max clauses returned. The full spec has ~3000 clauses; the default is a safe slice.
- `section` (string): Filter to clauses whose section number starts with this prefix, e.g. '22.2' for RegExp or '15' for the Locale-aware operations in ECMA-402.
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `clause.outline` (~203 tokens)

Spec section outline

Return the section tree (table of contents) for a parsed (spec, edition). `depth` caps tree depth (1 = top-level only). `under` anchors at a specific clause id so you get just its descendants. Each node carries { id, number, title, kind, children }.

Input parameters:

- `depth` (integer): Max tree depth to return. 1 = top-level only; 2 = first two levels; omitted = full tree.
- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).
- `under` (string): Optional clause id. If set, return only descendants of this clause.

### `spec.search` (~214 tokens)

Search spec

Search the parsed spec by clause id / aoid / title (and step text when search_steps is true). Returns lightweight hits ranked by match quality — the entry point when you don't know the exact clause id. `spec` selects '262' or '402'. Follow up with clause.get.

Input parameters:

- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `limit` (integer): Max ranked hits returned.
- `query` (string, required): Search text. Matched against clause id, aoid, and title (and step text when search_steps is true).
- `search_steps` (boolean): Also match against algorithm step text. Slower + noisier; off by default.
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `spec.crossrefs` (~348 tokens)

Spec cross-references

For a clause id, return its outgoing references (clauses it cites) and/or incoming references (clauses that cite it — the back-reference index the parse alone doesn't expose). Direction: 'in' | 'out' | 'both' (default). Outgoing also carries an `external` category: the clause's citations to external specs (Unicode, IETF, WHATWG) as resolvable URLs. Set `include_cross_spec: true` to also resolve outgoing references from ECMA-262 → ECMA-402 (or vice versa).

Input parameters:

- `direction` (string): 'in' = clauses that reference this one (back-refs); 'out' = clauses this one references; 'both' = both.
- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `id` (string, required): Spec clause id, e.g. 'sec-tonumber' (262) or 'sec-intl.numberformat' (402).
- `include_cross_spec` (boolean): If true, outgoing references also include AOIDs that resolve to the *other* TC39 spec (262 ↔ 402). Useful for queries like 'every 262 op that calls into Intl'. Off by default because it requires load…
- `limit` (integer): Max hits returned in each direction (incoming and outgoing are limited independently).
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `spec.diff` (~166 tokens)

Diff spec editions

Clause-level diff across any two editions of one spec. Defaults: from='latest', to='main' (working draft). Reports status (identical / modified / added / removed / missing-from-both) plus a field-level diff: title, signature, step count, per-step reworded indices, notes, crossrefs. `spec` selects '262' or '402'.

Input parameters:

- `from` (string): The 'before' edition. Defaults to the latest stable release.
- `id` (string, required): Spec clause id.
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).
- `to` (string): The 'after' edition. Defaults to the working draft (main).

### `spec.history` (~192 tokens)

Clause git history

Recent commits in the vendored spec checkout that touched a clause's `id="..."` token. Uses git pickaxe (`-S`) so it catches clause creation, deletion, and edits to the opening tag reliably; interior-text-only edits won't show. Returns SHA, date, author, subject per commit. `spec` selects '262' or '402'.

Input parameters:

- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `id` (string, required): Spec clause id.
- `limit` (integer): Max commits returned from the vendored spec checkout's git log.
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `spec.symbol_resolve` (~216 tokens)

Resolve spec symbol

Resolve spec notation like `[[Prototype]]` (internal slot), `%Object.prototype%` (well-known intrinsic), or `~number~` (sigil enum) to the clauses that mention or define it. Hits ranked by occurrence count + section-prefix bumps for the canonical definition location. `spec` selects '262' or '402'.

Input parameters:

- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `limit` (integer): Max candidate clauses returned, ranked by relevance score.
- `notation` (string, required): Spec notation like `[[Prototype]]` (internal slot), `%Object.prototype%` (well-known intrinsic), or `~number~` (sigil enum).
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `spec.sdo_index` (~250 tokens)

Syntax-directed operations index

Index Syntax-Directed Operations by the grammar production they handle. SDOs are abstract operations (Evaluation, BoundNames, etc.) with one `<emu-alg>` per production. Default by='production' returns { [production]: [{ sdo, id, title }] }; by='sdo' returns { [sdo title]: [productions] }. `filter` substring-narrows keys; `spec` selects '262' or '402'.

Input parameters:

- `by` (string): Index direction. 'production' (default) groups SDO definitions by the production they handle. 'sdo' groups productions by which SDO defines them.
- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `filter` (string)
- `limit` (integer): Cap the number of groups returned. Each group can still hold many entries.
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `spec.global_search` (~146 tokens)

Search both specs

Run spec.search across both ECMA-262 and ECMA-402 in one call and interleave results by score. Each hit is tagged with the spec it came from. Useful when you don't know which spec defines the symbol (e.g. `Canonicalize` is 262, `CanonicalizeLocaleList` is 402).

Input parameters:

- `limit` (integer): Total hits across both specs combined.
- `query` (string, required): Search text. Matched against clause id, aoid, and title across both specs (and step text when search_steps is true).
- `search_steps` (boolean): Also match against algorithm step text. Slower + noisier; off by default.

### `spec.well_known_intrinsics` (~209 tokens)

Well-known intrinsics

Enumerate the well-known intrinsics (`%X%` notations) used in the spec, with each one's probable defining clause (chosen by a title-substring heuristic — see `matched_on` per hit). For the canonical 262 well-known intrinsics table, read `clause.get { id: 'sec-well-known-intrinsic-objects' }` directly.

Input parameters:

- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `filter` (string): Case-insensitive substring filter on the intrinsic name (bare, e.g. 'object.prototype').
- `limit` (integer): Max well-known intrinsics returned.
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `spec.tables` (~249 tokens)

Spec tables

List or fetch parsed `<emu-table>` content. Pass `id` to get one table with full columns + rows; omit `id` to list tables (lightweight summaries) optionally filtered by caption/id substring. Authoritative source for the well-known intrinsics table (id='table-well-known-intrinsic-objects'), well-known symbols, completion record fields, etc. `spec` selects '262' or '402'.

Input parameters:

- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `filter` (string): Case-insensitive substring filter on the caption or id (list mode only).
- `id` (string): If set, return exactly this table (full columns + rows). If omitted, list tables (lightweight rows).
- `limit` (integer): Max table summaries returned in list mode (ignored when 'id' is set).
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `spec.grammar` (~284 tokens)

Spec grammar productions

Query standalone `<emu-grammar>` productions from the spec's lexical / syntactic grammar (§11-15 in 262). Three modes: { nonterminal: 'X' } returns every production for X; { contains: 'Y' } returns productions whose RHS or non-terminal name contains Y; neither returns a list of all non-terminals with their production counts. Set include_sdo:true to also surface SDO-attached grammar headers.

Input parameters:

- `contains` (string): Filter to productions whose RHS lines or non-terminal name contain this substring (case-insensitive).
- `edition` (string): Edition within the chosen spec. ECMA-262: es2016 … es2026, main. ECMA-402: es2016 … es2026, main. Aliases: latest, draft, next.
- `include_sdo` (boolean): If true, also include productions captured as SDO algorithm headers. Off by default — most callers want the standalone lexical/syntactic grammar definitions.
- `limit` (integer): Max productions (or non-terminal groups in list mode) returned.
- `nonterminal` (string): Filter to productions defining this non-terminal (exact match). Example: 'BindingIdentifier'.
- `spec` (string): Which TC39 spec to read: '262' (core language, default) or '402' (Internationalization API).

### `test262.search` (~253 tokens)

Search test262

Search tc39/test262 for tests matching a free-text query and/or an esid (clause id, prefix-matched). test262 covers both ECMA-262 and ECMA-402. Served from a parsed test262 index sourced via the loader chain (local cache → hosted Worker → bundled fallback); the index is fetched on first use and cached locally. If no layer can produce the index the result is empty + a hint explaining the one-time local build. No auth, no subprocess.

Input parameters:

- `esid` (string): Filter to tests whose front-matter esid: starts with this prefix (case-insensitive). Prefix match catches the common case where test262 uses a more specific esid than the spec section id — e.g. `esid…
- `limit` (integer): Max ranked hits returned from the test262 index.
- `query` (string): Free-text query. Matched case-insensitively as whitespace-separated tokens (AND) across each test's description + path. Either `query` or `esid` (or both) must be supplied.

### `test262.get` (~129 tokens)

Get test262 fixture

Fetch one test's source + parsed front-matter by path within the vendored tc39/test262 checkout. Pairs with test262.search — the paths it returns plug in here directly. Returns { source, front_matter, test262_sha, url } or { hint } if the path can't be resolved.

Input parameters:

- `path` (string, required): Path within the test262 checkout, relative to the repo root. Example: 'test/built-ins/Number/prototype/toString/S15.7.4.2_A1_T01.js'. The values returned by test262.search go here directly.

### `proposal.list` (~254 tokens)

List TC39 proposals

List TC39 proposals from a parsed proposals index sourced via the loader chain (local cache → hosted Worker → bundled fallback); the index is fetched on first use and cached locally. Filter by `stage` ('0'|'1'|'2'|'2.7'|'3'|'finished'|'inactive'|'active'), `champion` (substring), or `contains` (name/slug substring). Returns lightweight rows; follow up with `proposal.get`. No auth, no subprocess.

Input parameters:

- `champion` (string): Case-insensitive substring filter on the champion list.
- `contains` (string): Case-insensitive substring filter applied to the proposal name + slug.
- `limit` (integer): Max proposals returned from the filtered set.
- `spec` (string): Filter to one spec's proposals: '262' (core language) or '402' (Intl). tc39/proposals tracks the two in parallel — omit to list both.
- `stage` (string): Filter to one stage: '0', '1', '2', '2.7', '3', 'finished', 'inactive', or 'active' (anything in the active README — stages 2/2.7/3).

### `proposal.get` (~99 tokens)

Get TC39 proposal

Fetch one TC39 proposal by slug (exact) or name (case-insensitive). Returns { slug, name, stage, authors, champions, url, test262_flag, source_file }. Slug is canonical — use what proposal.list returns directly.

Input parameters:

- `name` (string, required): Match against either the proposal's slug (preferred, exact-match) or its name (case-insensitive). Slug is the canonical id — use what proposal.list returns directly.

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/xyzzylabs-tc39-mcp/tc39-mcp-2#diagnostics

## Score history

- 2026-08-03: 81
- 2026-08-02: 80
- 2026-08-01: 31
- 2026-07-31: 5
- 2026-07-30: 35
- 2026-07-28: 53
- 2026-07-27: 53

## Links

- npm package: https://www.npmjs.com/package/tc39-mcp
- Socket report: https://socket.dev/npm/package/tc39-mcp
- Repository: https://github.com/xyzzylabs/tc39-mcp
- Website: https://mcp.xyzzylabs.ai/tc39
- Changelog RSS feed: https://verifymcp.io/servers/xyzzylabs-tc39-mcp/tc39-mcp-2/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/xyzzylabs-tc39-mcp/tc39-mcp-2/changelog.json
- HTML version of this page: https://verifymcp.io/servers/xyzzylabs-tc39-mcp/tc39-mcp-2