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/madamak/apache-airflow-mcp-server/agents-mdgit clone --depth 1 https://github.com/madamak/apache-airflow-mcp-serverWrote 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/madamak/apache-airflow-mcp-server/agents-md)<a href="https://agentmods.dev/instructions/madamak/apache-airflow-mcp-server/agents-md"><img src="https://agentmods.dev/badge/instructions/madamak/apache-airflow-mcp-server/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 | $0.01534 | $0.01534 |
| Opus 5 | $0.00767 | $0.00767 |
| Sonnet 5 | $0.00307 | $0.00307 |
| Haiku 4.5 | $0.00153 | $0.00153 |
Grade A, and why
apache-airflow-mcp-server AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for AI Agents working on the Airflow MCP Server. This file orients you at the package level (what belongs where, how to extend safely). For module-level guidance, see src/airflow_mcp/AGENTS.md. For tests, see tests/AGENTS.md.
Project Layout
src/airflow_mcp/server.py: FastMCP entrypoint. Registers tools and HTTP routes. Thin wrappers only. Write tools are registered through thewrite_tooldecorator, which is a no-op in read-only mode.tools/: Business logic for MCP tools (structured dict outputs; FastMCP serializes). No FastMCP or web concerns. One module per domain (dags.py,runs.py,tasks.py,task_logs.py,datasets.py,instances.py); shared helpers in_common.py.registry.py: Instance registry loader (YAML with${VAR}env substitution, or a single instance built fromAIRFLOW_MCP_HOST/credentials env vars) and cached accessor.client_factory.py: Builds cachedapache-airflow-clientclients per instance; applies auth, SSL, timeouts. Handles both client majors: instances withapi_version: v1use the 2.x codegen (airflow_client.client.apis),api_version: v2uses the 3.x codegen (top-level API classes, JWT auth via/auth/token).get_api_family(instance)is the branch point tools use for version-specific calls.url_utils.py: URL resolver/builder (instance detection, dag/task/run extraction, UI URL construction).validation.py: Input validation (safe identifier patterns; SSRF guard via host check in resolver).observability.py: Structured logging with request_id; common operation logger.
errors.py: Error types andhandle_errorsdecorator (raises MCPToolErrorwith compact JSON payload).config.py: Pydantic settings (AIRFLOW_MCP_*).
tests/: Unit and integration tests (mocked client; no real network).examples/instances.yaml: Example instance registry used by tests.
Development Workflow
- Install & sync deps with
uv sync. - Run locally:
uv run airflow-mcp --transport stdiouv run airflow-mcp --transport http --host 127.0.0.1 --port 8765
- Test & lint before pushing:
uv run pytestuv run ruff check .
- Optional format pass:
uv run ruff format . - Commit hygiene: run tests + lint locally prior to 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.
- 5d ago First seen · 85 lines · 1,534 tokens per session scan A 0903b04a5ffa
apache-airflow-mcp-server AGENTS.md is an instructions file published in the GitHub repository madamak/apache-airflow-mcp-server (2 stars, last pushed 11d ago), licensed Apache-2.0. It adds 1,534 tokens to every session, about $0.0077 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
mcp-email-server AGENTS.md
AGENTS.md instructions for Wh1isper/mcp-email-server, covering repository overview, project conventions, current architecture direction, development workflow and documentation requirements.
RivalSearchMCP AGENTS.md
AGENTS.md instructions for damionrashford/RivalSearchMCP, covering agents.md, project overview, development commands, setup & running and install dependencies.
RivalSearchMCP CLAUDE.md
Claude Code instructions for damionrashford/RivalSearchMCP, covering claude.md, project overview, development commands, setup & running and set up pre-commit hooks.
mcp-outline CLAUDE.md
Instructions for Vortiago/mcp-outline, covering mcp outline server guide, purpose, architecture, tool categories and feature registration flow.
unraid CLAUDE.md
Claude Code instructions for dinglebear-ai/unraid, covering claude.md, what this is, identity and history (read before touching any url), "unraid-mcp" means five different things and agent plugins ship no claude code hooks.
blender-mcp AGENTS.md
AGENTS.md instructions for sandraschi/blender-mcp, covering blender-mcp agent context and quick ref.