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/detailobsessed/efficient-gitlab-mcp/claude-mdgit clone --depth 1 https://github.com/detailobsessed/efficient-gitlab-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/detailobsessed/efficient-gitlab-mcp/claude-md)<a href="https://agentmods.dev/instructions/detailobsessed/efficient-gitlab-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/detailobsessed/efficient-gitlab-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 | $0.00941 | $0.00941 |
| Opus 5 | $0.00470 | $0.00470 |
| Sonnet 5 | $0.00188 | $0.00188 |
| Haiku 4.5 | $0.00094 | $0.00094 |
Grade A, and why
efficient-gitlab-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 3d 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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Efficient GitLab MCP
An enhanced fork of zereight/gitlab-mcp with progressive disclosure pattern for token-efficient GitLab server management.
Quick Reference
bun install # Install dependencies
bun run build # Build: src/server/index.ts → dist/
bun test # Run all tests (Bun test runner)
bun run check # Lint + format (Biome, auto-fix)
bun run dev # Build and start
Architecture
src/
├── server/
│ ├── index.ts # MCP server entry point (stdio + streamable-http)
│ └── config.ts # ServerConfig interface + loadConfig()
├── tools/ # One file per GitLab domain (15 modules)
│ ├── issues.ts
│ ├── merge-requests.ts
│ ├── pipelines.ts
│ └── ...
├── registry/
│ ├── disclosure.ts # SDK-native progressive disclosure (enable/disable)
│ └── categories.ts # Tool category definitions
└── utils/
├── gitlab-client.ts # GitLab API client (singleton, supports graphql())
└── logger.ts # Structured MCP protocol logger
Progressive Disclosure
Tools start disabled (via toolRef.disable()). The LLM discovers them via meta-tools (list_categories, activate_tools), then enables categories on demand. Activation sets tool.enabled directly and sends a single batched notifications/tools/list_changed.
Adding a New Tool
- Add the handler in the appropriate
src/tools/<category>.tsfile - Use Zod for input schema validation
- Register with
server.registerTool(), thentoolRef.disable()(starts hidden) - Add the tool name to the category's tool map
- Add tests in
tests/<category>.test.ts
Code Style
- Biome with strict rules:
noExplicitAny,noNonNullAssertion,noExcessiveCognitiveComplexity - Spaces for indentation, double quotes
- Run
bun run checkto auto-fix
Testing
- Tests live in
tests/*.test.ts(17 test files) - Bun's built-in test runner
- Tests mock the GitLab API client — no live API calls
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.
- 3d ago First seen · 96 lines · 941 tokens per session scan A 7b2d86f8d182
efficient-gitlab-mcp CLAUDE.md is an instructions file published in the GitHub repository detailobsessed/efficient-gitlab-mcp (3 stars, last pushed 3mo ago), licensed MIT. It adds 941 tokens to every session, about $0.0047 per session on Opus 5. A static security scan graded it A with 0 findings. It comes from a forked repository.
Other instructions, from other repositories
gini-agent AGENTS.md
AGENTS.md instructions for Open-Curiosity/gini-agent, covering gini agent instructions, shape, adrs, boundaries and branches.
bread CLAUDE.md
Instructions for tjakoen/bread, covering claude.md — start here, what this is, start here (reading order), this repo is the stack's control plane — run it and non-negotiables (see conventions for the full rules).
agentconfig.org AGENTS.md
Instructions for agentconfig/agentconfig.org, covering agent instructions for agentconfig.org, project overview, target audience, site structure and tech stack.
rosetta command-versions-vs-presence.instructions.md
Instructions for tikoci/rosetta, a project described as: MCP Server with RouterOS docs + commands + products + changelogs, using SQLite-as-RAG, sourced from MikroTik.
titen CLAUDE.md
Instructions for RamaAditya49/titen, covering claude.md, read agents.md first, commands, two runtimes, one core and publishing to npm.
sandy CLAUDE.md
Instructions for jamestelfer/sandy, covering claude.md, voice, what is sandy, project layout and dev commands.