Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/umputun/local-docs-mcp/claude-mdgit clone --depth 1 https://github.com/umputun/local-docs-mcpWrote 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.
[](https://agentmods.dev/instructions/umputun/local-docs-mcp/claude-md)<a href="https://agentmods.dev/instructions/umputun/local-docs-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/umputun/local-docs-mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.01483 | $0.01483 |
| Opus 5 | $0.00741 | $0.00741 |
| Sonnet 5 | $0.00297 | $0.00297 |
| Haiku 4.5 | $0.00148 | $0.00148 |
Grade A, and why
local-docs-mcp CLAUDE.md 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 254 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Local Docs MCP Server - Development Guide
Project Overview
Go implementation of Model Context Protocol (MCP) server providing Claude with access to local markdown documentation from multiple sources.
Original implementation: ~/.dot-files/claude/local-docs-mcp.py
Architecture
Package Responsibilities
app (main)
- entry point and initialization
- logging setup
- server lifecycle management
- stdio transport configuration
app/server
- MCP server creation and configuration
- tool registration with official SDK
- server startup and shutdown
app/tools
- tool implementations (search_docs, read_doc, list_all_docs)
- input/output type definitions
- fuzzy matching logic
- business logic isolated from MCP protocol
app/scanner
- file discovery across sources (commands, project-docs, project-root)
- safe path resolution preventing traversal
- file metadata collection (size, source)
- directory walking with filtering rules
Design Principles
Type Safety
- use structs with JSON tags for all tool inputs/outputs
- leverage Go's type system for compile-time safety
- no interface{} except where SDK requires it
Error Handling
- return errors to caller, never panic
- wrap errors with context using fmt.Errorf
- consistent error response format for MCP tools
Testing
- minimum 80% coverage
- table-driven tests for all functions
- integration tests with real file fixtures
- separate unit tests from integration tests
Security
- safe path resolution mandatory for all user inputs
- no symlink following outside base directories
- reject absolute paths
- validate file sizes before reading
- UTF-8 validation on file content
Dependencies
Core
github.com/modelcontextprotocol/go-sdk/mcp- official MCP SDK
Utilities
github.com/sahilm/fuzzyorgithub.com/lithammer/fuzzysearch/fuzzy- fuzzy matchinggithub.com/stretchr/testify- testing assertions
Optional
github.com/go-pkgz/lgr- logging (or use standard log)
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.
- 5d ago First seen · 254 lines · 1,483 tokens per session scan A 129b782d5a1a
local-docs-mcp CLAUDE.md is an instructions file published in the GitHub repository umputun/local-docs-mcp (21 stars, last pushed 17d ago), licensed MIT. It adds 1,483 tokens to every session, about $0.0074 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 2026-08-30.
Other instructions, from other repositories
mcp-pandoc AGENTS.md
Instructions for vivekVells/mcp-pandoc, covering agents.md, what this is, where things are, deeper context and commands.
agentic-resume-builder CLAUDE.md
Claude Code instructions for WtotdeD/agentic-resume-builder, covering agentic-resume-builder development guidelines, what this repo is, active technologies, package manager and validation.
fcp-sheets CLAUDE.md
Instructions for os-tack/fcp-sheets, covering fcp-sheets, project overview, architecture, key patterns and commands.
rodya-caijing-studio AGENTS.md
AGENTS.md instructions for nekopunch11/rodya-caijing-studio, covering rodya-caijing-studio · 给 ai agent 的使用说明, 入口与路由, 不可破的硬约束（合规灵魂，破了整套工具就失效）, docx 渲染路由（财经内容台例外） and 在 codex 中安装.
paperless-ngx-mcp-server CLAUDE.md
Claude Code instructions for cbsmiley/paperless-ngx-mcp-server, covering claude.md, project overview, common commands, architecture and entry points.
hermes-procurement-pricing-mcp AGENTS.md
AGENTS.md instructions for felix-windsor/hermes-procurement-pricing-mcp, a project described as: Secure procurement quotation comparison MCP service with document parsing, mock market pricing, anomaly detection, and Telegram/Hermes integration.