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.
git clone --depth 1 https://github.com/s977043/river-reviewWrote 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/commands/s977043/river-review/register-plugin-asset)<a href="https://agentmods.dev/commands/s977043/river-review/register-plugin-asset"><img src="https://agentmods.dev/badge/commands/s977043/river-review/register-plugin-asset.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 5.1 | $0.00028 | $0.01587 |
| Opus 5 | $0.00014 | $0.00794 |
| Sonnet 5 | $0.00006 | $0.00317 |
| Haiku 4.5 | $0.00003 | $0.00159 |
Grade A, and why
register-plugin-asset scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
新しい配布アセット（command / agent / agent-skill）を plugin manifest に登録し、検証まで通す。手順の SSoT は docs/development/plugin-asset-registration-checklist.md にあり、本コマンドはその実行手順の具体化のみを担う。$ARGUMENTS で種別と名前を受け取る（省略時は git status の差分から種別を推定する）。
Step 0. 配布面か非配布面かを判定
git status --short
- 追加ファイルが
commands/*.md・agents/*.md・skills/agent-skills/**のいずれかを確認する .claude/commands/*.mdはリポ開発専用で配布されない。この場合 manifest 登録は不要と報告して終了する（skills/{upstream,midstream,downstream}/も pluginskillsの対象外＝別系統）- 判定基準の詳細は checklist の「配布面と非配布面の区別」表を正とする
Step 1. 種別ごとの manifest 登録
command（commands/<name>.md）
.claude-plugin/plugin.json の commands[] に "./commands/<name>.md" を追加する。.codex-plugin/plugin.json に commands フィールドはないため対応不要。commands/ 直下にはコマンド本体以外（README 等）を置かない（公式 validator が全 *.md をコマンドとして走査するため）。
- CLAUDE.md「Custom Commands」表と
docs/development/distributed-commands.mdに説明を追記する（npm run check:doc-enumが両表とcommands/*.mdの一致を機械検証するため、条件付きではなく必須）
agent（agents/<name>.md）
.claude-plugin/plugin.json の agents を確認する。現在は単一文字列（"./agents/river-review.md"）のため、2つ目を足すときは配列化する:
"agents": ["./agents/river-review.md", "./agents/<name>.md"]
- 単一文字列のまま新 agent を足すと解決されない。plugin-manifest スキーマ（
$schema）が配列を許容することを Read で確認する
agent-skill（skills/agent-skills/<name>/SKILL.md）
両 manifest の skills はディレクトリ参照（"./skills/agent-skills/"）のため、manifest 編集は不要。ディレクトリ配置だけで自動包含される。
- ディレクトリ名が kebab-case で frontmatter
metadata.nameと一致することを Read で確認する
Step 2. フィールドを追加・変更した場合の追加対応
manifest に新しい「フィールド」を足した／値を変えた場合（ファイル追加だけでなく）:
- 外部 bundle（awesome-codex-plugins fork の
plugin.json）にも同じ変更を反映する（CLAUDE.md「Plugin bundle mirror」）。反映は同 PR で行う package.jsonSSoT の同期フィールド（keywords / homepage / author / license）は手編集せず、npm run plugin:syncで反映する
Step 3. 検証シーケンス
npm run plugin:validate
npm run plugin:sync:check
npm run agent-skills:validate # agent-skill を触った場合のみ
npm run check:doc-enum
npm run meta:validate
- 全て pass を確認する。fail は該当項目を修正してから再実行する
plugin:validateが「参照パス不在」で落ちる＝登録漏れ or パス誤り。plugin:sync:checkの drift＝同期フィールドの手編集を疑うcheck:doc-enumが落ちる＝docs/development/distributed-commands.mdか CLAUDE.md「Custom Commands」表への追記漏れ（meta:validateからも実行されるため CI で必ず止まる）
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.
- 8d ago First seen · 88 lines · 28 tokens per session scan A e74d4f1b4a92
register-plugin-asset is a command published in the GitHub repository s977043/river-review (3 stars, last pushed today), licensed MIT. It adds 28 tokens to every session and 1,587 once invoked, about $0.0001 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 commands, from other repositories
logic-diff
Check two code versions for semantic equivalence — use after a refactor or rewrite.
logic-review
Review code for logic bugs — use when you suspect something is wrong but have no failing test yet.
consensus-review
Consensus code review (Opus arbiter + codex, optional pi advisor) across 4 dimensions (architecture/quality/impact/tests) for a PR or uncommitted changes; P0–P2 findings; strict sceptic by default; every flag also accepted with a leading -.
architecture-audit
One-shot architecture audit — layers, slices, contracts, paradigms, complexity.
arch-review
Architectural impact review of a change (diff-scoped) via the arch-reviewer agent.
impact-review
Review of a change and its impact on adjacent parts via the impact-reviewer agent.