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 skills/nwiizo/tfmcp/releasenpx skills add nwiizo/tfmcp --skill releasegit clone --depth 1 https://github.com/nwiizo/tfmcpWhat 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 | $0.00040 | $0.00465 |
| Opus 5 | $0.00020 | $0.00233 |
| Sonnet 5 | $0.00008 | $0.00093 |
| Haiku 4.5 | $0.00004 | $0.00047 |
Grade A, and why
release 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.
What it actually says
Release tfmcp
Treat publication as an irreversible operation. Keep the version synchronized
across Cargo.toml, Cargo.lock, server.json, the OCI identifier, and
Dockerfile.
Prepare
- Confirm the target version does not already exist on crates.io, as a Git tag, or as a GitHub release.
- Update the synchronized version fields and add a dated
CHANGELOG.mdsection. - Update the tested Terraform version in CI and both Docker stages when the release changes the supported baseline.
- Run
./Release.sh vX.Y.Z. Fix every failure; do not weaken gates to make a release pass. - Review
cargo package --listand the generated package contents for secrets, temporary files, and stale release documents.
Publish
- Commit the verified release preparation on
main. - Push
mainand confirm the Rust CI workflow succeeds for that exact commit. - Re-run
./Release.sh vX.Y.Zfrom the clean commit. - Run
./Release.sh vX.Y.Z --publish. - Verify the crates.io version, annotated Git tag, public GitHub release, multi-platform GHCR image, and GitHub Actions results.
Never publish from a dirty worktree, bypass a failing check, reuse a release version, or move an existing release tag.
Required gates
The release script is authoritative. It runs formatting, Clippy with warnings
denied, locked tests, audit, cargo coupling, similarity-rs, metadata
validation, diff checks, release build, package verification, and a crates.io
publish dry run.
Use cargo coupling --ai --git-months 6 --exclude-tests to diagnose structural
hotspots and similarity-rs src --skip-test --threshold 0.90 --min-lines 8 to
inspect duplicate candidates. Refactor findings only when the new boundary or
abstraction has a clear owner and preserves behavior.
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 · 48 lines · 40 tokens per session scan A 08f01c6560d1
release is a skill published in the GitHub repository nwiizo/tfmcp (371 stars, last pushed 4d ago), licensed MIT. It adds 40 tokens to every session and 465 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
review-pr
Use when reviewing a pull request - security-focused review following repo agent guidance for breaking changes, malicious patterns, and backward compatibility.
upgrade-cluster
Use when upgrading or hardening an existing kube-hetzner cluster, including module version bumps, provider lockfile refreshes, k3s/RKE2 channel/version upgrades, immutable node replacement, system-upgrade-controller changes, or live cluster rollout validation.
fix-issue
Use when working on a GitHub issue - fetches issue details, analyzes codebase, implements fix following project methodology.
sync-docs
Use when documentation needs updating - ensures variables.tf, docs/llms.md, kube.tf.example, and README are in sync.
triage-issue
Use when triaging a GitHub issue - analyzes issue, checks for duplicates, categorizes, and drafts response.
auditing-terraform-infrastructure-for-security
Auditing Terraform infrastructure-as-code for security misconfigurations using Checkov, tfsec, Terrascan, and OPA/Rego policies to detect overly permissive IAM policies, public resource exposure, missing encryption, and insecure defaults before cloud deployment.