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/nrflo/nrflo/do-releasenpx skills add nrflo/nrflo --skill do-releasegit clone --depth 1 https://github.com/nrflo/nrfloWhat 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 | $0.00089 | $0.02692 |
| Opus 5 | $0.00044 | $0.01346 |
| Sonnet 5 | $0.00018 | $0.00538 |
| Haiku 4.5 | $0.00009 | $0.00269 |
Grade A, and why
do-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 2d 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 — 227 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/do-release
Cuts a new nrflo release end-to-end. Invoked as:
/do-release patch
/do-release minor
/do-release major
The argument is required. If missing or not one of patch|minor|major, stop and tell the user the valid options.
Step 0 — Confirm the pre-release gate has been run
Before doing anything else, ask the user whether they've already run /pre-release against the commits about to be tagged. The /pre-release skill runs the full manual_testing harness across every supported provider × mode combination and is the only gate that exercises the real CLI binaries against real provider credentials. Releases that skip it have historically shipped regressions that no Go test could have caught.
Prompt verbatim:
Have you already run
/pre-releasefor the commits you're about to tag? (yes / no)
- yes → ask the user to paste the VERDICT line from that run. If they reply
VERDICT: PASS, proceed to Step 1. If anything else (PASS with caveats, FAIL, or "I lost it"), stop and tell them to re-run/pre-releaseand come back when it's clean. - no → stop. Tell them to run
/pre-releasefirst, and only invoke/do-releaseonce it reportsVERDICT: PASS.
Do not offer to run /pre-release yourself from inside this skill — it takes ~15–20 minutes (gated by the slowest folder, normally engine at ~13 min) and the user should be aware they're committing to that wall time. They invoke it explicitly.
What you're automating
- nrflo version lives in two places that must stay in lock-step:
VERSIONfile at repo root (plainMAJOR.MINOR.PATCH, nov). Source of truth for humans and Makefile.- Git tag
vMAJOR.MINOR.PATCH. Source of truth for GoReleaser ({{ .Version }}).
- Pushing a
v*tag triggers two workflows in parallel:.github/workflows/release.yml→ GoReleaser → GitHub Release + Homebrew tap atnrflo/homebrew-tap(darwin binaries)..github/workflows/docker.yml→ multi-arch build →ghcr.io/${owner}/nrflo-server:${VERSION}+:latest(linux/amd64,arm64, api-mode).
- GoReleaser writes an auto-generated body. You overwrite it with a richer, human-readable summary via
gh release edit. - The container image tag drops the
vprefix (e.g.v0.3.6→ghcr.io/nrflo/nrflo-server:0.3.6). This is enforced by theIMAGE_TAGrule in the Makefile and bydocker/metadata-actionindocker.yml.
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.
- 2d ago First seen · 227 lines · 89 tokens per session scan A de54cf4f3436
do-release is a skill published in the GitHub repository nrflo/nrflo (2 stars, last pushed 21d ago), licensed Apache-2.0. It adds 89 tokens to every session and 2,692 once invoked, about $0.0004 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 skills, from other repositories
codew-release-qa-sweep
Use before claiming Codewhale release work is done: run the full gate sweep and list the manual QA targets.
gh-credit-harvest
Harvest one community PR into a release branch with authorship and credit preserved, verified green, and a warm thank-you.
release
Prepare a named version: preflight, version consistency, build/package, smoke test, checksums/notes, and release readiness. Publishing/tagging/deploy need separate authorization. Explicit-only.
gh-find-prs
Survey open Codewhale PRs and triage each for mergeability and disposition against the real landing branch.
best-of-n
Generate a small set of independent candidate solutions in worktrees, judge them against one explicit rubric, and apply the winner only after PASS verification.
contributor-onboarding
Help a new contributor get productive on this checkout - inspect sync state against main, build, run the repository's exact verification gate, and produce a local what's-new digest. Never fetches, pulls, or modifies a dirty tree on its own. Explicit-only.