kaji: Skill for Claude Code

.claude/skills/release-starter/SKILL.md

release-starter is a skill for Claude Code from apokamo/kaji. It costs 34 tokens per session (1,556 once invoked), scanned A, original, Apache-2.0.

A publishing workflow for a managed starter template snapshot. It resolves the tracking issue and candidate, requires an independent review pass, checks repository and dependency consistency, and then performs an atomic push after its approval gates.

In plain words
What is it for?
Use it with a tracking issue ID when publishing a verified starter template snapshot, including cases where the candidate has no changes or contains changes.
Why use it?
It prevents an unreviewed, stale, mismatched, or unverified starter snapshot from being published.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is apokamo/kaji's own configuration. It tells Claude Code how to work on kaji itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything kaji configures →

Reuse

Borrowing it

Nothing to install: this file belongs to apokamo/kaji. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/apokamo/kaji/main/.claude/skills/release-starter/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/apokamo/kaji

Made for: Claude Code.

Wrote 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.

agentmods badge for release-starter

README.md
[![agentmods](https://agentmods.dev/badge/skills/apokamo/kaji/release-starter.svg)](https://agentmods.dev/skills/apokamo/kaji/release-starter)
Your own site
<a href="https://agentmods.dev/skills/apokamo/kaji/release-starter"><img src="https://agentmods.dev/badge/skills/apokamo/kaji/release-starter.svg" alt="Measured on agentmods" height="20"></a>
Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,556 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00034 $0.01556
Opus 5 $0.00017 $0.00778
Sonnet 5 $0.00007 $0.00311
Haiku 4.5 $0.00003 $0.00156

Measured 4d ago against content hash 95d6475054a8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

release-starter 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.

.claude/skills/release-starter/SKILL.md · 92 lines

How it starts

The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Release Starter

/release-starter <tracking_issue_id> で managed starter の検証済み template snapshot を公開する。

Pre-flight

  1. tracking Issue と starter sync runbook から target(= 本文の現在の syncing batch から求まる active_target)、covered_targets(同じ batch の target 集合)、starter identity、local path を解決する。同じ本文読み取りから tracking_issue_has_pending_tasks(現在の batch 以外に open 行が残っているか)も導出する。
  2. kaji issue resolve-verdict <id> --step review-starter-update --require-meta target --require-meta base --require-meta candidate を実行し、最新 verdict が独立 review PASS で、 meta.target == active_target、meta.candidate == local main HEAD であることを確認する。未検出、 不正 marker、meta 欠落、不一致は fail-closed で ABORT。
  3. local main clean、target kaji Release published、dependency / lockfile version 整合、quality gate を確認する。
  4. meta.base == meta.candidate の N/A を release-plan より先に分岐し、N/A では release-plan を呼ばない。ただし N/A でも close 前に remote main == meta.base (== meta.candidate) を必須とし、review PASS 後に remote main が前進していれば stale review evidence として ABORT する。変更 candidate の場合だけ、この時点で初めて pre-flight and recovery を読み、手順1で導出した tracking_issue_has_pending_tasks を含む観測 JSON を release-plan へ渡す。未公開 path は remote main == meta.base、公開済み残処理 path は remote main == meta.candidate を必須にする。

Publish

  • N/A (meta.base == meta.candidate): 上記の先行分岐で、独立 review PASS かつ remote main == meta.base (== meta.candidate) を確認した後だけ完了 bookkeeping(後述)を実行する。 remote main が前進していれば stale review evidence として ABORT する。starter tag / Release は 作らない。
  • ref push path: candidate SHA と kaji-vX.Y.Z または kaji-vX.Y.Z-rN を提示し、workflow 外で 人間の明示承認を得てから annotated tag と main を git push --atomic する。
  • push 後は Release notes template から gh release create を実行し、 続けて完了 bookkeeping を行う。
  • route 2 の部分成功再実行は release-plan の remaining_actions のうち create_release だけを 実行する(不足していれば gh release create を再試行。新 tag、ref push、再承認は不要)。 update_state_table / close_tracking_issue / promote_next_task は状態表更新・Issue close が未完了であることを示す参考情報に留め、直接実行しない。実行は必ず下記の完了 bookkeeping(kaji starter task-plan)経由で行う(判定の二重化防止。詳細は pre-flight and recovery)。

Read the full file on GitHub · 92 lines

Files

What ships with it

2 files 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.

Changes

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.

  1. 4d ago Changed · +42 lines 95d6475054a8
  2. 8d ago First seen · 50 lines · 34 tokens per session scan A 48ccbebdd506

Subscribe to this mod's changes

release-starter is a skill published in the GitHub repository apokamo/kaji (12 stars, last pushed today), licensed Apache-2.0. It adds 34 tokens to every session and 1,556 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.