FastMCP is a Python framework for building Model Context Protocol applications, which connect language models with tools and data through servers and clients. Developers use it to expose Python functions as tools, connect to local or remote MCP servers, and build interactive tool interfaces. The catalogue includes eight skills, three instructions, two rules, one setting, and one hook for FastMCP.
Borrowing it
Nothing to install: this file belongs to PrefectHQ/fastmcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/PrefectHQ/fastmcp/main/.agents/skills/release/SKILL.mdgit clone --depth 1 https://github.com/PrefectHQ/fastmcpWrote 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/skills/prefecthq/fastmcp/release)<a href="https://agentmods.dev/skills/prefecthq/fastmcp/release"><img src="https://agentmods.dev/badge/skills/prefecthq/fastmcp/release/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/prefecthq/fastmcp/release"><img src="https://agentmods.dev/badge/skills/prefecthq/fastmcp/release.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- Snyk pass
- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium MCP Rug Pull · line 55 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
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 | $0.00083 | $0.03337 |
| Opus 5.5 | $0.00033 | $0.01335 |
| Sonnet 5.5 | $0.00017 | $0.00667 |
| Haiku 4.5 | $0.00008 | $0.00334 |
Grade A, and why
release scanned grade A with 1 finding 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 today.
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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Confirm each package with `curl -fsS https://pypi.org/pypi/<pkg>/<new>/json | jq .info.version`; How it starts
The opening of the file, as written. The whole thing — 251 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cutting a release
A release is a tag on the branch that owns the line: <branch> below is main for
the current major and release/3.x for 3.x maintenance. Every step runs against that
branch, including the docs PR, so a maintenance entry lands in the maintenance
branch's own changelog above its newest 3.x entry. The tag triggers Publish fastmcp-slim to PyPI;
its success fans out to fastmcp-tasks, fastmcp-remote, and fastmcp, and the
fastmcp publish job opens the published-docs PR for stable release-triggered
publication when the release SHA is an ancestor of the default branch. Prereleases
and manual package-workflow dispatch do not open that PR. Mintlify builds docs/
from published-docs at https://gofastmcp.com/docs; the separate
PrefectHQ/gofastmcp-website repository deploys its main to Vercel for the landing
page and /docs proxy. Two hand-maintained docs files mirror every release and must be on the
tagged commit, so the docs PR always lands first. PyPI releases are immutable; a bad
one gets a follow-up patch, never a re-tag.
Procedure
-
Preview what's in it.
git fetch origin && git log v<prev>..origin/<branch> --format='%h %s' gh api -X POST repos/PrefectHQ/fastmcp/releases/generate-notes \ -f tag_name=v<new> -f target_commitish=<branch> -f previous_tag_name=v<prev> --jq .body \ > /tmp/generated-notes.md uv run scripts/release_contributor_credit.py /tmp/generated-notes.md \ > /tmp/contributor-credit.jsonRead the two most recent releases for voice:
gh release list -L 5, thengh release view <tag>on each.Review the scan's attribution candidates. The helper batch-fetches metadata only for PRs in the generated notes. It suggests shared credit for human authors of the PR commits and final merged commit, plus issue reporters whose issues are closed by a maintainer PR, even when their co-author trailer is missing. Human co-authors are included even when the PR has no linked issue. It excludes known agent identities and bots; a GitHub
Useridentity alone does not prove someone is human. Maintainer permission checks are reused across PRs.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- today Changed · +34 lines c8935eacfa69
- 1mo ago Changed · +56 lines ae98e73c9763
- 1mo ago Changed 450402dbf841
- 1mo ago Changed 50c4079acddc
- 2mo ago Changed · +1 lines 8babb30cf7af
- 2mo ago First seen · 160 lines · 83 tokens per session scan A ad6164e836b2
release is a skill published in the GitHub repository PrefectHQ/fastmcp (28,322 stars, last pushed today), licensed Apache-2.0. It adds 83 tokens to every session and 3,337 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
release compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| releasePrefectHQ/fastmcp | Skill | 28k | today | Apache-2.0 | A | 83 | Claude Code, Codex |
| agent-framework-py-releasemicrosoft/agent-framework | Skill | 15k | yesterday | MIT | A | 102 | Claude Code, Codex |
| release-reviewrohitg00/ai-engineering-from-scratch | Skill | 77k | yesterday | MIT | A | 7 | Claude Code, Codex |
| orchard-releaserohitg00/ai-engineering-from-scratch | Skill | 77k | yesterday | MIT | A | 11 | Claude Code, Codex |
| orchestratehumanlayer/humanlayer | Skill | 13k | yesterday | Apache-2.0 | A | 49 | Claude Code |
| setup-worktreehumanlayer/humanlayer | Skill | 13k | yesterday | Apache-2.0 | B | 16 | Claude Code |
| create-milestoneagentic-community/mcp-gateway-registry | Skill | 1.0k | yesterday | Apache-2.0 | A | 64 | Claude Code |