agentmods.dev

nacre: Skill for Claude Code

.claude/skills/mcp-tool/SKILL.md

mcp-tool is a skill for Claude Code from nacre-work/nacre. It costs 110 tokens per session (827 once invoked), scanned A, original, Apache-2.0.

A set of rules for building and changing MCP tools. MCP is a standard way for AI agents to call external tools and services.

In plain words
What is it for?
Use it when adding MCP tools, changing how they connect, listing available tools, or updating OAuth and authorization behavior.
Why use it?
It keeps tool requests, permissions, errors, and HTTP or command-line connections consistent. This reduces accidental data exposure and deployment problems.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is nacre-work/nacre's own configuration. It tells Claude Code how to work on nacre itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything nacre configures →

not rated 0repo 8d ago A scan 110 tokens original Apache-2.0
View source ↗ nacre-work/nacre nacre.work
Reuse

Borrowing it

Nothing to install: this file belongs to nacre-work/nacre. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/nacre-work/nacre/main/.claude/skills/mcp-tool/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/nacre-work/nacre

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for mcp-tool

README.md
[![agentmods](https://agentmods.dev/badge/skills/nacre-work/nacre/mcp-tool.svg)](https://agentmods.dev/skills/nacre-work/nacre/mcp-tool)
Your own site
<a href="https://agentmods.dev/skills/nacre-work/nacre/mcp-tool"><img src="https://agentmods.dev/badge/skills/nacre-work/nacre/mcp-tool.svg" alt="Measured on agentmods" height="20"></a>
Per session 110 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 827 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00110 $0.00827
Opus 5 $0.00055 $0.00413
Sonnet 5 $0.00022 $0.00165
Haiku 4.5 $0.00011 $0.00083

Measured 6d ago against content hash 668e1b90e651, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

mcp-tool scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/mcp-tool/SKILL.md · 77 lines

How it starts

The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Adding or changing an MCP tool

Contract: docs/mcp.md, target revision 2026-07-28.

Transport rules that constrain every tool

  • Stateless. No initialize, no Mcp-Session-Id, nothing kept between requests. Any request is served by any replica.
  • A tool needing state between calls returns an explicit descriptor in its result and takes it back as an argument next time. Hidden state in the transport is not allowed — it is what makes the round-robin deployment work.
  • Required headers: MCP-Protocol-Version, Mcp-Method, Mcp-Name.
  • tools/list returns ttlMs: 300000 and cacheScope: "user". The catalog depends on the caller's permissions, so a global cache would serve one user's catalog to another.

Every tool declares

  1. Which permission it requires — read or write, from the model in docs/authz.md. Remember write does not imply read: an ingest-only service account must not be able to search.
  2. What it returns on no permission — 404, never 403, and worded identically to a genuinely missing object.

Forbidden in a tool

  • Error messages that reveal an inaccessible object exists. "Layer contracts not found" and "you may not read layer contracts" must be the same string.
  • Accepting org_id as an argument. It comes from the token, always.
  • Bypassing the authorization service on the search path "for speed".

Descriptions are generated, not written

search builds its description from the caller's visible layer catalog:

Search corporate documents by meaning and by exact term — identifiers, error codes, part numbers and names match literally. Available: {layer.name} — {layer.description} ({n} docs); …

A generic "searches the knowledge base" makes the model reach for web search instead. The layer description column is user-facing copy for this reason — treat it as product text, not an internal note.

Because the description depends on permissions, the tool list is per-user. This is the same fact as cacheScope: "user" above; if you change one, check the other.

Read the full file on GitHub · 77 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 77 lines · 110 tokens per session scan A 668e1b90e651

Subscribe to this mod's changes

mcp-tool is a skill published in the GitHub repository nacre-work/nacre (0 stars, last pushed 8d ago), licensed Apache-2.0. It adds 110 tokens to every session and 827 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Backend and APIsBuilding agentsmcpself-hostedsemantic-searchvector-database
Related

Other skills, from other repositories

chroma

Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source…

ihatesea69/HieuNghi-AI-Skills · 63 tokens

chroma

Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source…

davila7/claude-code-templates · 63 tokens

chroma

Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source…

synthetic-sciences/openscience · 63 tokens

chroma

Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source…

OpenLAIR/dr-claw · 63 tokens

chroma

Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source…

Orchestra-Research/AI-Research-SKILLs · 63 tokens

chroma

Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source…

Microck/ordinary-claude-skills · 63 tokens