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/blackaxgit/ssh-mcp/agents-mdgit clone --depth 1 https://github.com/blackaxgit/ssh-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 | $0.05785 | $0.05785 |
| Opus 5 | $0.02892 | $0.02892 |
| Sonnet 5 | $0.01157 | $0.01157 |
| Haiku 4.5 | $0.00579 | $0.00579 |
Grade A, and why
ssh-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 yesterday.
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 — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
AI coding assistant context for this repository. Read natively by Codex, Cursor, Copilot, Gemini CLI, Windsurf, and 20+ AGENTS.md-aware tools. Claude Code users: load via
@AGENTS.mdinCLAUDE.md.
Command sources: .github/workflows/ci.yml (the gates), .github/workflows/release.yml (build/publish), CONTRIBUTING.md, pyproject.toml. There is no Makefile — copy commands from CI rather than inventing them.
Overview
- Purpose: SSH MCP server that lets AI assistants execute commands on remote servers.
- Owner: TODO: team/maintainer
- Type: app — published as a PyPI wheel (
uvx blc-ssh-mcp) and a container image (ghcr.io/blackaxgit/ssh-mcp). The PyPI distribution isblc-ssh-mcp, notssh-mcp— that name belongs to an unrelated project this repo has never owned; see the comment abovenameinpyproject.toml. The import package is stillssh_mcpand the image is still.../ssh-mcp. - Stacks: Python 3.11-3.14,
uv+ hatchling, asyncssh, MCP SDK, Starlette/uvicorn (HTTP transport only) - Deploy targets: N/A — this is a tool operators run themselves, not a hosted service
- Version:
src/ssh_mcp/__init__.pyis the single source (hatchling reads it). Currently0.5.6;CHANGELOG.md[Unreleased]targets 0.6.0.
What this tool does matters for how you treat it. Every tool call runs a shell command or transfers a file on a remote host. A bug here is not a crashed request; it is an unintended command on someone's infrastructure, or a write to the operator's own machine.
Commands
uv sync --locked --extra dev # setup. --locked is required; see Gotchas
uv run pytest # full suite (~690 tests, seconds)
uv run pytest tests/test_ssh.py::TestRedactSecrets -v # one class
uv run pytest 'tests/test_ssh.py::TestRedactSecrets::<test_name>' -v # one test
uv run pytest -k "confinement" -v # by keyword
Full gate, exactly as CI runs it — run all six before opening a PR:
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.
- yesterday First seen · 198 lines · 5,785 tokens per session scan A 7142c86c7e62
ssh-mcp AGENTS.md is an instructions file published in the GitHub repository blackaxgit/ssh-mcp (1 stars, last pushed 27d ago), licensed MPL-2.0. It adds 5,785 tokens to every session, about $0.0289 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
spiceai copilot-instructions.md
Instructions for spiceai/spiceai, covering spice.ai agent instructions, data correctness — absolute top priority, evidence — no claim without a reproduction, build, test, lint (expensive — read first) and git & prs.
spiceai AGENTS.md
Instructions for spiceai/spiceai, a project described as: Add a real-time analytics node to your operational database. Spice is a portable, accelerated SQL query, search, and LLM-inference engine in Rust for data-grounded AI apps and agents.
Cotal AGENTS.md
Instructions for Cotal-AI/Cotal, covering agents.md, what this is, read these first, commands and repository map.
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.
agentarea AGENTS.md
Instructions for agentarea/agentarea, covering agents.md, where to look, structure, commands and backend (from agentarea-platform/).
mcp-server-kubernetes CLAUDE.md
Instructions for Flux159/mcp-server-kubernetes, covering claude.md, development commands, build and development, testing and quality and local development testing.