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/j7an/nexus-mcp/agents-mdgit clone --depth 1 https://github.com/j7an/nexus-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/j7an/nexus-mcp/agents-md)<a href="https://agentmods.dev/instructions/j7an/nexus-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/j7an/nexus-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 | $0.01449 | $0.01449 |
| Opus 5 | $0.00724 | $0.00724 |
| Sonnet 5 | $0.00290 | $0.00290 |
| Haiku 4.5 | $0.00145 | $0.00145 |
Grade A, and why
nexus-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 — 200 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - Guidelines for AI Coding Agents
Project Overview
Nexus MCP is a Python 3.13+ MCP server that enables AI models to invoke agent runners and integrations (Claude Code, Codex, OpenCode, OpenCode server) as tools. Built with FastMCP 3.1+.
OpenCode Gemini-family model names are OpenCode provider/model configuration, not support for a separate command-line runner.
Build, Lint, Test Commands
# Setup (first time)
uv sync
uv run pre-commit install
# Run all tests (unit + e2e, fast)
uv run pytest
# Run single test file
uv run pytest tests/unit/test_types.py
# Run single test function
uv run pytest tests/unit/test_types.py::test_prompt_request_valid
# Run with coverage (threshold: 90%)
uv run pytest --cov=nexus_mcp -v
# Integration tests (slow - real CLI binaries, run before PR only)
uv run pytest -m integration
# Type check
uv run mypy src/nexus_mcp
# Lint
uv run ruff check .
# Format
uv run ruff format .
# Auto-fix lint issues
uv run ruff check --fix .
# Run all pre-commit hooks
uv run pre-commit run --all-files
Branch and Commit Guidance
- Never commit to
mainunless explicitly directed. - For issue or feature work, create a branch or worktree from
origin/main, keep all changes there, and prepare the result as a PR. - Never commit spec or plan documents unless explicitly asked.
- Do not use
git add -Aorgit add .; stage explicit paths. - Ask before adding new dependencies or making destructive changes.
Release Workflow Guidance
- Keep PyPI and TestPyPI Trusted Publishing jobs caller-owned in
.github/workflows/release.yml. Do not replace them withj7an/shared-workflows/.github/workflows/publish-pypi.yml; PyPI does not authorize cross-repo reusable workflows as trusted-publisher workflows. - When porting fixes from
shared-workflows, use the caller-owned PyPI Trusted Publishing template and preserve Nexus-specific safeguards:VERIFY_PYTHON: "3.13",VERIFY_COMMAND: nexus-mcp --version,scripts/derive-published-version.shbefore upload, TestPyPI and PyPI environment URLs, and MCP Registry publishing gated on GitHub release success. - Keep
setup-uvcaching disabled in the TestPyPI verification job while it intentionally skips checkout and creates the throwaway.verifyproject later. If caching is re-enabled, the job must first provide real dependency files forsetup-uvto hash so release runs do not emit empty-workdir or cache-dependency warnings. - Pin reusable
shared-workflowscallers to immutable commit SHAs with a version comment, but keep PyPI publishing inline unless PyPI explicitly supports cross-repo reusable workflows for Trusted Publishing.
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 · 200 lines · 1,449 tokens per session scan A f1d3878224b8
nexus-mcp AGENTS.md is an instructions file published in the GitHub repository j7an/nexus-mcp (4 stars, last pushed yesterday), licensed MIT. It adds 1,449 tokens to every session, about $0.0072 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
agent-drop GEMINI.md
Gemini CLI instructions for mikezio/agent-drop: When this repo is configured as an MCP server in a CLI agent.
mooc-starter-kit AGENTS.md
AGENTS.md instructions for jamditis/mooc-starter-kit, covering agents.md, one-shot runs, worked example and more.
mooc-starter-kit GEMINI.md
Gemini CLI instructions for jamditis/mooc-starter-kit, covering gemini.md, free tier, one-shot runs, worked example and more.
mooc-starter-kit CLAUDE.md
Claude Code instructions for jamditis/mooc-starter-kit, covering claude.md, beat, style, source standards and terminology.
aoc-utils-mcp GEMINI.md
Gemini CLI instructions for derailed-dash/aoc-utils-mcp, covering advent-of-code (aoc) utils, aoc-utils mcp server and general instructions.
claude-tui AGENTS.md
AGENTS.md instructions for slima4/claude-tui, covering agents.md, quick commands, testing, gotchas (hard-won knowledge) and local development.