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/nguyenchiencong/local-docs-mcp/agents-mdgit clone --depth 1 https://github.com/nguyenchiencong/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/nguyenchiencong/local-docs-mcp/agents-md)<a href="https://agentmods.dev/instructions/nguyenchiencong/local-docs-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/nguyenchiencong/local-docs-mcp/agents-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 | $0.00733 | $0.00733 |
| Opus 5 | $0.00367 | $0.00367 |
| Sonnet 5 | $0.00147 | $0.00147 |
| Haiku 4.5 | $0.00073 | $0.00073 |
Grade A, and why
local-docs-mcp AGENTS.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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Overview
This is a modular semantic search system with MCP (Model Context Protocol) integration for indexing and searching local documentation using Qdrant vector database and Ollama embeddings. The system provides AI assistants with semantic search capabilities through the Model Context Protocol.
Key Commands
Indexing Documents
# Index documents from the docs/ directory
uv run python -m src.indexing.main_flow
# Alternative using the project script
local-docs-index
Running the MCP Server
# Start the MCP server for AI assistant integration
uv run python -m src.mcp_server.server
# Alternative using the project script
local-docs-mcp
Testing
# Run tests with pytest directly (if available)
uv run pytest tests/
Architecture Overview
Core Components
-
Indexing Module (
src/indexing/)main_flow.py: Main CocoIndex flow for document processing and embedding generationchunking.py: Text chunking utilities using Chonkie library
-
Search Service (
src/search/)service.py: Decoupled semantic search service with various search strategiesmodels.py: Data models for search results and configuration
-
MCP Server (
src/mcp/)server.py: MCP server implementation exposing search toolstools.py: MCP tool definitions and handlers
Configuration
- Environment Variables: Set in
.envfile (copy from.env.example) - Project Settings:
pyproject.tomlcontains dependencies and build configuration
Key Dependencies
- cocoindex: Document processing and indexing framework
- qdrant-client: Vector database client
- ollama: Local AI model serving
- mcp: Model Context Protocol library
- chonkie: Text chunking library
Development Notes
Document Indexing
- Documents are sourced from the
docs/directory by default - Text is chunked using Chonkie library before embedding
- Embeddings are generated using Ollama with the specified model
- Results are stored in Qdrant collection named "local-docs-collection"
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 · 102 lines · 733 tokens per session scan A b041cda3b3b0
local-docs-mcp AGENTS.md is an instructions file published in the GitHub repository nguyenchiencong/local-docs-mcp (2 stars, last pushed 9mo ago), licensed MIT. It adds 733 tokens to every session, about $0.0037 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.
Other instructions, from other repositories
PixelRAG CLAUDE.md
Instructions for StarTrail-org/PixelRAG, covering pixelrag, layout and conventions.
OpenDocuments AGENTS.md
Instructions for joungminsung/OpenDocuments, covering agents.md, project overview, monorepo structure, common commands and 특정 패키지만.
pdf-inline-image-rag-mcp AGENTS.md
AGENTS.md instructions for Joncallim/pdf-inline-image-rag-mcp, covering pdf inline image rag mcp agent guide, invariants, repository map, focused workflow and validation.
doctree-mcp CLAUDE.md
Claude Code instructions for joesaby/doctree-mcp, covering claude.md — doctree-mcp, project overview, architecture, key design decisions and data flow.
google-drive-rag-mcp AGENTS.md
AGENTS.md instructions for asamiile/google-drive-rag-mcp, covering google drive rag mcp — agent instructions, project overview, key files, running the server and install dependencies.
MCP-Google-Doc CLAUDE.md
Claude Code instructions for ophydami/MCP-Google-Doc, covering claude.md, build & run, architecture, current state and target architecture: vector rag support tool.