Przejdź do treści
Biblioteka skilli
AI i agenci · Przegląd i audyt

audit

/audit

Audytuje definicje narzędzi MCP pod kątem jakości: opisy parametrów, adnotacje i limity rozmiaru wyników.

Licencja
MIT
Ostatnia zmiana repozytorium
26.10.2026

Ocena i bezpieczeństwo

84/ 100Klasa: dobry
Format9 / 25
Frontmatter z nazwą i opisem, poprawna nazwa, opis do 1024 znaków, nazwa zgodna z folderem.
Odkrywalność25 / 25
Opis mówi, kiedy użyć skilla, ma sensowną długość i jest w 3. osobie.
Treść35 / 35
Co najmniej 150 słów i do 500 linii, nagłówki, listy kroków, przykłady albo kod.
Higiena15 / 15
Bez zaślepek typu TODO, bez ścieżek Windows, z licencją.
Ryzykobrak sygnałów w skanie
Pliki w paczce
12
Skrypty
2
Komentarze HTML
0
Adresy URL w SKILL.md
5
Przestarzałość

Brak sygnałów. Repozytorium zmienione 26.10.2026, bez odwołań do wycofanych modeli i API.

Licencja

MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.

Co sprawdzić przed instalacją
  • 2 skrypty w paczce. Przejrzyj je przed instalacją, bo Claude może je uruchomić.

Skrypty: scripts/discover.sh, scripts/discover.test.sh

Co robi i kiedy

Audytuje definicje narzędzi MCP pod kątem jakości: opisy parametrów, adnotacje i limity rozmiaru wyników.

Wyzwalacze z opisu

audit MCP toolscheck MCP tool descriptionsreview MCP server qualitytool annotationsreadOnlyHint missingparameter descriptions missingcheck the _meta annotationsmaxResultSizeChars

Argumenty

/audit [path]
Oryginalny opis (EN)

Audit MCP server tool definitions against design quality criteria. Use when: 'audit MCP tools', 'check MCP tool descriptions', 'review MCP server quality', 'tool annotations', 'readOnlyHint missing', 'parameter descriptions missing', 'check the _meta annotations', 'maxResultSizeChars', 'requiresUserInteraction', 'alwaysLoad', 'are my server instructions too long', 'mcp audit', or before shipping MCP server changes. Optional path argument targets a single server directory; omit to audit the whole project. Produces per-tool PASS/WARN/FAIL scorecard covering description completeness, parameters,

Jak działa

  1. Purpose
  2. Server discovery configuration
  3. Arguments
  4. Track progress
  5. Workflow
  6. What this skill does NOT do
Sekcje z nagłówków SKILL.md, w oryginale · 133 linii, 971 słów
Podgląd SKILL.md
---
description: "Audit MCP server tool definitions against design quality criteria. Use when: 'audit MCP tools', 'check MCP tool descriptions', 'review MCP server quality', 'tool annotations', 'readOnlyHint missing', 'parameter descriptions missing', 'check the _meta annotations', 'maxResultSizeChars', 'requiresUserInteraction', 'alwaysLoad', 'are my server instructions too long', 'mcp audit', or before shipping MCP server changes. Optional path argument targets a single server directory; omit to audit the whole project. Produces per-tool PASS/WARN/FAIL scorecard covering description completeness, parameters, naming, annotations, and the Claude Code `_meta` annotations, plus a server-level result for the server `instructions` size budget. Language-agnostic: Python (`mcp`), TypeScript, .NET. Not for: whether configured MCP servers are safe to run (/mcp-tools:audit-posture), or MCP config correctness and connection issues (/harness-config:audit)."
argument-hint: "[path]"
user-invocable: true
disable-model-invocation: false
context: fork
background: false
metadata:
  workflow-stage: review
  summary: Audit MCP tool definitions against design quality criteria
---

**Arguments.** `[path]`. A directory to scope the audit to (e.g. a single server dir), or omit for the whole project.

## Purpose

Evaluate MCP server tool definitions against design quality criteria drawn from three upstream
authorities, cited (not recapped) so the current text always governs:

- [MCP specification 2026-07-28: Tools](https://modelcontextprotocol.io/specification/2026-07-28/server/tools). The normative protocol (MUST / SHOULD / OPTIONAL requirements for names, schemas, annotations); why this revision is in reference/checklist.md, "Spec revision record".
- [Define tools: best practices for tool definitions](https://platform.claude.com/docs/en/agents-and-tools/tool-use/define-tools#best-practices-for-tool-definitions). Tool-design guidance for descriptions, parameters, namespacing, and workflow-shaped granularity (correlate with [Anthropic: Writing effective tools for AI agents](https://www.anthropic.com/engineering/writing-tools-for-agents)).
- [Claude Code: Connect Claude Code to tools via MCP](https://code.claude.com/docs/en/mcp). Claude-Code-specific client behavior: `_meta` annotations and result-size limits. The dated pointer records for the values C4 and C17-C19 turn on are in reference/checklist.md, "Client-behavior record" and "Description length record".

Produces a per-tool scorecard with actionable findings. Catches description gaps, missing annotations,
and naming issues before they degrade LLM tool selection accuracy.

This skill runs as a **blocking fork** (`context: fork`, `background: false`; what that changes is
owned by the
[invocation-context rubric](https://github.com/melodic-software/claude-code-plugins/blob/main/docs/conventions/invocation-context/README.md)).
`$ARGUMENTS` (an optional path) and the working tree are the whole input.

## Server discovery configuration

Phase 1 enumerates MCP servers and their tool source files by scanning the project for the per-language
tool markers. The `tool-marker`, `name-extraction`, `description-extraction`, and `meta-extraction`
rules are upstream MCP-SDK conventions, stable across repos. See
[reference/server-discovery.md](reference/server-discovery.md) for the per-language discovery rules and
how servers are grouped.

## Arguments

Parse `$ARGUMENTS`:

- **`<path>`**. Audit a single scope. A directory to narrow the scan to (typically one server's directory).
- ***(empty)***. Audit every tool discovered under the project root.

## Track progress

For multi-server audit runs (Phases 1-3 across ≥2 servers), keep an in-response checklist of the three
phases and tick each as it completes. Phase 2 may run subagent fan-out for ≥5 tools; check each
subagent's evidence before accepting its verdicts, then consolidate them into the one Phase 3 report.

## Workflow

### Phase 1: Discover servers and tools

Run `bash "${CLAUDE_PLUGIN_ROOT}/skills/audit/scripts/discover.sh"`, adding `--path <dir>` when
`$ARGUMENTS` supplies a directory, to scope the scan to that directory.

### Phase 2: Evaluate against checklist

Read each `Tool file:` from Phase 1. Per the language rules in
[reference/server-discovery.md](reference/server-discovery.md), extract descriptions, parameters,
wire-level annotations (C12-C14), and the tool's `_meta` object (C17-C19). The last via
**meta-extraction**, recording each key's JSON type and not merely its presence, because C18 turns on
it. Also record each parameter schema's `x-mcp-header` value, if any (C20). Load the detailed checklist from [reference/checklist.md](reference/checklist.md).

Once per server, also resolve its `instructions` field. See **Server instructions** in
[reference/server-discovery.md](reference/server-discovery.md) and evaluate C4's per-server clause
against it. Phase 1's records are per-tool, so this is the only step that reaches it; its result lands
in the server-level row of the Phase 3 report, not in any tool's table.

Evaluate every criterion in the checklist against each tool, and C4's per-server clause once per
server. Record each result as:

- **PASS**. Criterion met
- **WARN**. Criterion partially met or could be improved
- **FAIL**. Criterion not met
- **info**. An optimization opportunity rather than a defect; the severity `reference/checklist.md`
  assigns to C8, C11, C14, and by default to C17-C19
- **n/a**. The criterion has no subject here, so it cannot pass: a server whose construction site
  declares no `instructions` gives C4's per-server clause nothing to size
- **undetermined**. The subject was not reachable in the scanned scope (no server construction site
  found), which is not the same as its being absent

### Phase 3: Report

Output a markdown report with this structure:

```markdown
# MCP Tool Audit Report

**Date:** YYYY-MM-DD
**Servers audited:** N
**Tools audited:** N
**Overall:** X pass, Y warn, Z fail, W info

## Summary by server

| Server | Tools | Pass | Warn | Fail | Info |
|--------|-------|------|------|------|------|
| <server-name> | N | ... | ... | ... | ... |

## Findings by server

### Server: <server-name> (<language>)

Server-level criteria, the outcomes that belong to the server rather than to any one tool:

| Criterion | Authority | Result | Details |
|-----------|-----------|--------|---------|
| C4 Server `instructions` within size budget | OPINION | n/a | Construction site declares no `instructions` |

#### Tool: <tool_name>

| Criterion | Authority | Result | Details |
|-----------|-----------|--------|---------|
| C1 Description has "what" | ANTHROPIC | WARN | Missing "when to use" context |
| C9 Name charset/length valid | SPEC-SHOULD | PASS | |
| C12 readOnlyHint set | SPEC-OPTIONAL | WARN | Read-only tool lacks the hint |
| C18 requiresUserInteraction is JSON `true` | OPINION | FAIL | Declared as the string `"true"`, so silently ignored |
| ... | ... | ... | ... |

(repeat for each tool)
```

**Prioritize FAIL items**. Highest-value improvements. WARN items are suggestions and info items are
optimizations. The `Overall` line and the summary table count every outcome recorded for a server:
its tools' criterion rows and its server-level rows, across the four severity buckets. `n/a` and
`undetermined` are not severities: they record that a criterion had no subject, or none reachable, so
they appear only in the server-level criterion table and never count as a pass. A missing annotation
is never FAIL. Annotations are
OPTIONAL in the spec (C12-C14) or Claude-Code-specific advisories (C17-C19); only a declared value
Claude Code silently ignores can FAIL (C18). Likewise an absent `x-mcp-header` passes, and only a
declared one that breaks the spec's constraints can FAIL (C20).

## What this skill does NOT do

- Does not modify tool definitions. It reports. Use findings to guide manual improvements.
- Does not test tool functionality. Use MCP Inspector for that.
- Does not evaluate MCP resources. Only tools.
- Does not judge whether the servers in your Claude Code configuration are safe to run (floating
  versions, publisher provenance, local stdio versus remote). Use `/mcp-tools:audit-posture` for that.
Pliki w paczce (12)
  • SKILL.md 8 KB
  • evals/evals.json 10 KB
  • evals/fixtures/meta-annotations.md 3 KB
  • evals/fixtures/over-length-description.md 7 KB
  • evals/fixtures/server-with-resource.md 1 KB
  • evals/fixtures/tools-with-defects.md 2 KB
  • evals/fixtures/well-designed-tool.md 1 KB
  • evals/fixtures/x-mcp-header.md 3 KB
  • reference/checklist.md 19 KB
  • reference/server-discovery.md 6 KB
  • scripts/discover.sh 8 KB
  • scripts/discover.test.sh 7 KB

Wersje

RepozytoriumWynikLicencja
Ta strona · melodic-software/claude-code-plugins84MIT
marcel-bich/marcel-bich-claude-marketplace100MIT
jeremylongshore/tons-of-skills-marketplace100MIT
friendlyinternet/nuxt-crouton100MIT
userKael/rastro100MIT
Archont561/qgis-agent-harness100MIT
lilly-logic-labs/lll-skills100MIT
Come-In-Labs/agent-skills100MIT
minanyang/10x-agent100MIT

Migawka bazy z 13.11.2026 · wersja paczki ca564777d5f1 (skrót SHA-256 całej paczki)

Pobrania w ostatnich 30 dniach: 0 · ranking

Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.

Podobne skille

  • 100tool-design-checker
    Ocenia definicje narzędzi serwera MCP tak, jak czyta je model: cel, opisy parametrów, adnotacje i rozmiar w tokenach.
  • 90audit-discoverability
    Sprawdza, czy agenci znajdą i wybiorą wtyczkę lub serwer MCP na podstawie opisów i odpowiedzi na typowe prompty.
  • 100mcp-tool
    Optymalizuje zestaw narzędzi zewnętrznego serwera MCP, gdy agent źle wybiera narzędzia lub błędnie wypełnia argumenty: opisy narzędzi, dokumentacja parametrów, przykłady i dobór eksponowanych narzędzi.
  • 100eval-mcp
    Mierzy, czy Claude poprawnie używa narzędzi serwera MCP: dokładność wyboru narzędzia, jakość schematów i iteracyjna optymalizacja opisów.
  • 100mcp-tool-naming-description-design
    Projektuje intuicyjne nazwy narzędzi, opisy bogate w wyzwalacze i kontrakty parametrów, by modele trafnie wybierały narzędzia.
  • 100tool-design
    Projektuje interfejsy narzędzi dla agentów LLM: opisy, schematy parametrów i komunikaty błędów, gdy model wybiera złe narzędzie.

Uwagi i zgłoszenia

Wiadomość trafi do Bartosza Sokolińskiego. Adres e-mail jest potrzebny tylko do odpowiedzi. Szczegóły w polityce prywatności.