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/code-yeongyu/senpi/release-publishnpx skills add code-yeongyu/senpi --skill release-publishgit clone --depth 1 https://github.com/code-yeongyu/senpiWrote 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/code-yeongyu/senpi/release-publish)<a href="https://agentmods.dev/skills/code-yeongyu/senpi/release-publish"><img src="https://agentmods.dev/badge/skills/code-yeongyu/senpi/release-publish.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 | $0.00044 | $0.01586 |
| Opus 5 | $0.00022 | $0.00793 |
| Sonnet 5 | $0.00009 | $0.00317 |
| Haiku 4.5 | $0.00004 | $0.00159 |
Grade A, and why
release-publish 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 4d 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Release Publish
Use this skill when releasing senpi with the canonical CalVer flow, pushing a release tag, publishing the GitHub Release, or publishing packages to npm.
Canonical release flow
Run scripts/release.mjs from a clean main checkout.
- Confirm you are on
main. - Confirm the worktree is clean.
- Confirm the release version is a valid CalVer.
- Run the changelog audit before release.
- Make only product-facing release notes from
packages/coding-agent/CHANGELOG.mdviascripts/release-notes.mjs. - Skip housekeeping, upstream-sync, and generated-catalog-only entries.
- Duplicate user-facing ai/agent/tui changes into the coding-agent changelog.
- Run
npm run check. - Run
npm run build. - Run
CI=1 npm test. - Create the release commit.
- Create the release tag.
- Create the next-cycle changelog commit.
- Push
mainand the tag.
Pre-flight checklist
- Read the current release rules in
AGENTS.mdandscripts/AGENTS.md. - Use a clean dedicated clone if the main checkout has foreign state.
- Never clean, stash, or otherwise disturb other people's work.
- Use the protected-main path when needed;
UPSTREAM_AUTOMATION_TOKENexists for authenticated protected-main pushes. - npm publish is delegated: the tag pipeline's
publish-npmjob no longer publishes — it dispatchespublish-npm.ymlin publish-only mode (the proven route from the v2026.7.28-x and v2026.8.4 recoveries) and gates the public GitHub Release on that run's success. npm's trusted publisher matches thepublish-npm.ymlworkflow identity only; never reintroduce a direct publish or anenvironment:on the publish path (the environment subject is what npm rejects). - Do not rerun
scripts/release.mjsafter the tag has been pushed. - If checks fail before the tag push, fix first, then re-release from the beginning.
- Treat
E404noise for@code-yeongyu/senpi-orchestratoras non-fatal. - Use
node scripts/release-notes.mjs/ the cl.md audit before publishing notes. - Keep release notes product-facing only.
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.
- 4d ago First seen · 93 lines · 44 tokens per session scan A 1be2850d836b
release-publish is a skill published in the GitHub repository code-yeongyu/senpi (419 stars, last pushed today), licensed MIT. It adds 44 tokens to every session and 1,586 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
pi
Use when needing a minimal, extensible terminal AI coding agent harness in TypeScript with plugin architecture. Pi: minimalist terminal AI coding agent emphasizing simplicity and composability.
speckit-archive
Archive a feature specification into main project memory after merge, resolving gaps and conflicts.
speckit-critique-run
Perform a dual-lens critical review of the specification and plan from both product strategy and engineering risk perspectives before implementation.
speckit-extract-run
Extract knowledge, guidelines, and ADRs from one or more completed spec directories into the project documentation system.
speckit-iterate-apply
Apply a pending iteration to spec documents — update all artifacts that speckit.implement relies on, then hand off to implementation.
speckit-auto-run
Run the full speckit workflow end-to-end — specify, plan, critique, tasks, implement, review, extract — making all decisions autonomously.