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/agents-mdgit clone --depth 1 https://github.com/detailobsessed/efficient-gitlab-mcpWhat 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 | $0.01790 | $0.01790 |
| Opus 5 | $0.00895 | $0.00895 |
| Sonnet 5 | $0.00358 | $0.00358 |
| Haiku 4.5 | $0.00179 | $0.00179 |
Grade A, and why
efficient-gitlab-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 2d 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 — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Conventions — efficient-gitlab-mcp
Project-specific conventions that complement CLAUDE.md. CLAUDE.md covers
architecture, build, and the progressive disclosure pattern; this file covers
workflow (branches, commits, tickets) and patterns (test idioms,
schema choices) used inside the codebase.
Verification commands
Always run before declaring a task done:
bun test # 220+ tests, ~500ms — fast enough to run on every change
bun run check # Biome lint + format, auto-fix
bun run build is run automatically by the pre-commit hook (via the
Build step). Do not skip the hook.
Git workflow
This repo uses git-spice (git-spice -h), not raw git for stack management.
- Trunk is
detailobsessed, NOTmain.mainis a read-only mirror ofzereight/gitlab-mcpupstream. - Stack branches are named
port/<short-kebab-description>(e.g.port/labels-as-array,port/get-file-contents-auto-ref). - Use
git-spice bc <name> -a -m "<msg>"to create a new branch on top of the current one, staging tracked changes in one step. - Use
git-spice ca -a --no-editto amend the current branch's commit. - Use
git-spice stack submit --update-onlyto push updates after amending. - Use
git-spice lsto visualise the stack (top of stack at top, trunk at bottom).
Commit conventions
Conventional commits with these prefixes:
feat(<scope>):new behaviour (tool, feature, capability)fix(<scope>):bug fixdocs(<scope>):documentation only (including doctring/description text)test(<scope>):test-only changeschore(<scope>):infrastructure / non-functional
Scope examples: tools, disclosure, schemas, users, repositories,
labels, mcp-server, projects, merge-requests.
The body should explain why, not just what. The bottom should always carry
a Closes DOT-NNN (or Refs DOT-NNN) line for Linear auto-close on merge.
Do not add Co-Authored-By: Claude / Devin / etc. trailers — ~/.claude/CLAUDE.md
disables the harness's auto-trailer and we don't want manual ones either.
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.
- 2d ago First seen · 165 lines · 1,790 tokens per session scan A 69621bba0a0e
efficient-gitlab-mcp AGENTS.md is an instructions file published in the GitHub repository detailobsessed/efficient-gitlab-mcp (3 stars, last pushed 3mo ago), licensed MIT. It adds 1,790 tokens to every session, about $0.0089 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
oh-my-pi AGENTS.md
AGENTS.md instructions for can1357/oh-my-pi, covering development rules, default context, package structure, github and code quality.
opengeni AGENTS.md
Instructions for Cloudgeni-ai/opengeni, covering agent / automation notes (opengeni), full local stack, architecture notes, pull-request delivery across moving main and keeping these notes current.
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).
copilot-money-mcp CLAUDE.md
Instructions for ignaciohermosillacornejo/copilot-money-mcp, covering copilot money mcp server, quick reference, architecture, data flow and project structure.
dunx CLAUDE.md
Instructions for petarzarkov/dunx, covering dunx - claude code instructions, rule 1 - native implementations only, never reimplement what bun already does, never invent what a mature library already solves and reuse the @arkv workspace - and extend it upstream.
steward AGENTS.md
Instructions for Steward-Fi/steward, covering repository operating guide, setup, validation, generated files and database migrations.