[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F1PXBY7QWN3V0P4TC1TWX6":3,"skill-same:01M7F1PXBY7QWN3V0P4TC1TWX6":46},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":29,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M7F1PXBY7QWN3V0P4TC1TWX6","review-agent-setup",null,"Configure human-in-the-loop gating for AI agent review actions in Claude Code. Use when setting up a project where an agent may post PR reviews, comments, merges, or edit CI configuration, and you want a cryptographically auditable approval trail with Cedar-enforced gates.","shotcowboystyle\u002Fag-skills","shotcowboystyle","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F62471?v=4","01M7EZTCFCZJ6K3MYW8BPM4X6R","Curtis Blanton","01M7EZTCG1S5M04J90A931X9BX",0,"MIT","https:\u002F\u002Fgithub.com\u002Fshotcowboystyle\u002Fag-skills\u002Fblob\u002Fmain\u002Fplugins\u002Freview-agent-governance\u002Fskills\u002Freview-agent-setup\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: review-agent-setup\ndescription: Configure human-in-the-loop gating for AI agent review actions in Claude Code. Use when setting up a project where an agent may post PR reviews, comments, merges, or edit CI configuration, and you want a cryptographically auditable approval trail with Cedar-enforced gates.\n---\n\n# review-agent-governance — Setup\n\nGate AI agent review actions (PR reviews, comments, merges, CI edits) behind\nexplicit human approval. Every attempt, approved or denied, produces an\nEd25519-signed receipt.\n\n## When to use this plugin\n\nInstall it in projects where a Claude Code agent:\n\n- Reviews, comments on, or merges pull requests (`gh pr review`, `gh pr merge`)\n- Triages issues (`gh issue comment`, `gh issue close`)\n- Publishes releases (`gh release create`)\n- Modifies CI configuration (`.github\u002Fworkflows\u002F`, `.gitlab-ci.yml`)\n- Pushes to protected branches (`main`, `master`, `release`, `production`)\n- Posts to external notification surfaces (Slack webhooks, Discord), once you\n  add a rule for the command that posts (the default policy does not gate them)\n\nIf the agent is only doing local file edits and running tests, this plugin is\noverkill. Use `protect-mcp` for general tool-call policy enforcement and skip\nthis one.\n\n## One-time setup\n\n### 1. Install the plugin\n\n```bash\nclaude plugin install shotcowboystyle\u002Fag-skills\u002Freview-agent-governance\n```\n\n### 2. Copy the default policy to your project\n\n```bash\ncp .claude\u002Fplugins\u002Freview-agent-governance\u002Fpolicies\u002Freview-agent-governance.cedar \\\n   .\u002Freview-governance.cedar\n```\n\nYou can edit this file to match your project's specific rules. See\n`..\u002Fagents\u002Freview-policy-author.md` for guidance on authoring review\npolicies.\n\n### 3. Create a receipts directory and sign key\n\n```bash\nmkdir -p .\u002Freview-receipts\necho \"\u002Freview-receipts\u002F\" >> .gitignore\necho \"\u002Freview-governance.key\" >> .gitignore\necho \"\u002F.review-approved\" >> .gitignore\nif [ ! -e .\u002Freview-governance.key ]; then\n  d=$(mktemp -d) && npx protect-mcp@0.7.4 init --dir \"$d\" && mv \"$d\u002Fkeys\u002Fgateway.json\" .\u002Freview-governance.key\nfi\n```\n\nprotect-mcp 0.7.4 `sign` does not create the key, so the last command creates\nit, and it never replaces an existing key. Without a key, the receipts are\nunsigned. To rotate the key, archive `.\u002Freview-governance.key` and\n`.\u002Freview-receipts\u002Freceipts.jsonl` first, then run the command again. Give auditors the `publicKey`\nvalue from `.\u002Freview-governance.key`. Do not commit the file, because it also\nholds the private key.\n\n## Per-session workflow\n\nThe Cedar policy denies review-surface actions unconditionally. To approve\na specific action, open an approval window before it and close it after.\n\n### Flag file (simplest)\n\n```bash\n# Before the action you want to approve\ntouch .\u002F.review-approved\n\n# Let Claude Code run the review \u002F comment \u002F merge\n\n# Immediately after\nrm .\u002F.review-approved\n```\n\n### Slash command (from within Claude Code)\n\n```\n\u002Fapprove-review \"Reviewing PR #123 authored by contributor X\"\n```\n\nThis creates `.\u002F.review-approved` with the given reason embedded as a note,\nand records the reason in an unsigned approval log under\n`.\u002Freview-receipts\u002Fapprovals\u002F`. A follow-up `rm` is still needed to close the\nwindow.\n\n### Dry-run everything (force full policy evaluation)\n\nIf you want every tool call to go through Cedar with no approval bypass:\n\n```bash\nexport REVIEW_APPROVAL_FLAG=.\u002F.never-approve\n```\n\nAny tool call matching a forbid rule will be denied; approved windows have\nno effect. Useful for CI or for a locked-down audit run.\n\n## Verifying the receipts\n\nList all receipts:\n\n```bash\nls -la .\u002Freview-receipts\u002F\n```\n\nVerify every receipt offline with the public key:\n\n```bash\nPUB=$(node -p 'JSON.parse(require(\"fs\").readFileSync(\".\u002Freview-governance.key\")).publicKey')\nnpx @veritasacta\u002Fverify@0.9.2 --replay-chain .\u002Freview-receipts\u002Freceipts.jsonl --key \"$PUB\"\n```\n\nExit 0 means every receipt verified. Exit 1 means a receipt failed\nverification, because it was tampered with, the key is wrong, or a line is\nmalformed. Exit 2 means the receipts file could not be read.\n\nA denied call never runs, so it has no receipt. To see what the policy\nblocked, run this inside Claude Code:\n\n```\n\u002Flist-pending\n```\n\nIt lists the tool calls that the PreToolUse hook blocked in the current\nsession, with the tool name and the command or path.\n\n## Example: approving a PR review\n\n```bash\n# 1. Human reviews the agent's proposed comment\n$ \u002Flist-pending\n  Blocked in this session:\n  - Bash \"gh pr review 42 --approve --body 'LGTM'\"\n  - Bash \"gh pr comment 42 --body 'Looking good'\"\n\n# 2. Human decides the first one is appropriate, approves it\n$ \u002Fapprove-review \"Approving LGTM on PR 42 after visual inspection\"\n  .\u002F.review-approved created\n\n# 3. Agent retries the action; this time it succeeds\n$ agent: gh pr review 42 --approve --body \"LGTM\"\n  [receipt appended to .\u002Freview-receipts\u002Freceipts.jsonl, decision=allow]\n\n# 4. Human closes the window\n$ rm .\u002F.review-approved\n```\n\nThe allowed call has a signed receipt that anyone with the public key can\nverify offline. The denied attempt has no receipt, and the approval log is\nnot signed, so keep both in mind when you show the trail to an auditor.\n\n## Composing with protect-mcp\n\nIf both plugins are installed, each plugin's `hooks\u002Fhooks.json` registers its\nown PreToolUse hook, and Claude Code runs both on every tool call:\n\n```json\n{ \"type\": \"command\", \"command\": \"\\\"${CLAUDE_PLUGIN_ROOT}\\\"\u002Fhooks\u002Fevaluate.sh\" }\n```\n\nEach `evaluate.sh` reads `tool_name` and `tool_input` from the hook payload on\nstdin (Claude Code sets no `TOOL_NAME` variable) and evaluates its own policy:\n`.\u002Fprotect.cedar` for protect-mcp and `.\u002Freview-governance.cedar` here.\n\nBoth hooks must pass for the tool call to proceed. Cedar deny in either\npolicy blocks it.\n\n## Standards\n\n- **Ed25519** — RFC 8032 (digital signatures)\n- **JCS** — RFC 8785 (deterministic JSON canonicalization)\n- **Cedar** — AWS's open authorization policy language\n- **IETF draft** — [draft-farley-acta-signed-receipts](https:\u002F\u002Fdatatracker.ietf.org\u002Fdoc\u002Fdraft-farley-acta-signed-receipts\u002F)\n","# review-agent-governance — Setup\n\nGate AI agent review actions (PR reviews, comments, merges, CI edits) behind\nexplicit human approval. Every attempt, approved or denied, produces an\nEd25519-signed receipt.\n\n## When to use this plugin\n\nInstall it in projects where a Claude Code agent:\n\n- Reviews, comments on, or merges pull requests (`gh pr review`, `gh pr merge`)\n- Triages issues (`gh issue comment`, `gh issue close`)\n- Publishes releases (`gh release create`)\n- Modifies CI configuration (`.github\u002Fworkflows\u002F`, `.gitlab-ci.yml`)\n- Pushes to protected branches (`main`, `master`, `release`, `production`)\n- Posts to external notification surfaces (Slack webhooks, Discord), once you\n  add a rule for the command that posts (the default policy does not gate them)\n\nIf the agent is only doing local file edits and running tests, this plugin is\noverkill. Use `protect-mcp` for general tool-call policy enforcement and skip\nthis one.\n\n## One-time setup\n\n### 1. Install the plugin\n\n```bash\nclaude plugin install shotcowboystyle\u002Fag-skills\u002Freview-agent-governance\n```\n\n### 2. Copy the default policy to your project\n\n```bash\ncp .claude\u002Fplugins\u002Freview-agent-governance\u002Fpolicies\u002Freview-agent-governance.cedar \\\n   .\u002Freview-governance.cedar\n```\n\nYou can edit this file to match your project's specific rules. See\n`..\u002Fagents\u002Freview-policy-author.md` for guidance on authoring review\npolicies.\n\n### 3. Create a receipts directory and sign key\n\n```bash\nmkdir -p .\u002Freview-receipts\necho \"\u002Freview-receipts\u002F\" >> .gitignore\necho \"\u002Freview-governance.key\" >> .gitignore\necho \"\u002F.review-approved\" >> .gitignore\nif [ ! -e .\u002Freview-governance.key ]; then\n  d=$(mktemp -d) && npx protect-mcp@0.7.4 init --dir \"$d\" && mv \"$d\u002Fkeys\u002Fgateway.json\" .\u002Freview-governance.key\nfi\n```\n\nprotect-mcp 0.7.4 `sign` does not create the key, so the last command creates\nit, and it never replaces an existing key. Without a key, the receipts are\nunsigned. To rotate the key, archive `.\u002Freview-governance.key` and\n`.\u002Freview-receipts\u002Freceipts.jsonl` first, then run the command again. Give auditors the `publicKey`\nvalue from `.\u002Freview-governance.key`. Do not commit the file, because it also\nholds the private key.\n\n## Per-session workflow\n\nThe Cedar policy denies review-surface actions unconditionally. To approve\na specific action, open an approval window before it and close it after.\n\n### Flag file (simplest)\n\n```bash\n# Before the action you want to approve\ntouch .\u002F.review-approved\n\n# Let Claude Code run the review \u002F comment \u002F merge\n\n# Immediately after\nrm .\u002F.review-approved\n```\n\n### Slash command (from within Claude Code)\n\n```\n\u002Fapprove-review \"Reviewing PR #123 authored by contributor X\"\n```\n\nThis creates `.\u002F.review-approved` with the given reason embedded as a note,\nand records the reason in an unsigned approval log under\n`.\u002Freview-receipts\u002Fapprovals\u002F`. A follow-up `rm` is still needed to close the\nwindow.\n\n### Dry-run everything (force full policy evaluation)\n\nIf you want every tool call to go through Cedar with no approval bypass:\n\n```bash\nexport REVIEW_APPROVAL_FLAG=.\u002F.never-approve\n```\n\nAny tool call matching a forbid rule will be denied; approved windows have\nno effect. Useful for CI or for a locked-down audit run.\n\n## Verifying the receipts\n\nList all receipts:\n\n```bash\nls -la .\u002Freview-receipts\u002F\n```\n\nVerify every receipt offline with the public key:\n\n```bash\nPUB=$(node -p 'JSON.parse(require(\"fs\").readFileSync(\".\u002Freview-governance.key\")).publicKey')\nnpx @veritasacta\u002Fverify@0.9.2 --replay-chain .\u002Freview-receipts\u002Freceipts.jsonl --key \"$PUB\"\n```\n\nExit 0 means every receipt verified. Exit 1 means a receipt failed\nverification, because it was tampered with, the key is wrong, or a line is\nmalformed. Exit 2 means the receipts file could not be read.\n\nA denied call never runs, so it has no receipt. To see what the policy\nblocked, run this inside Claude Code:\n\n```\n\u002Flist-pending\n```\n\nIt lists the tool calls that the PreToolUse hook blocked in the current\nsession, with the tool name and the command or path.\n\n## Example: approving a PR review\n\n```bash\n# 1. Human reviews the agent's proposed comment\n$ \u002Flist-pending\n  Blocked in this session:\n  - Bash \"gh pr review 42 --approve --body 'LGTM'\"\n  - Bash \"gh pr comment 42 --body 'Looking good'\"\n\n# 2. Human decides the first one is appropriate, approves it\n$ \u002Fapprove-review \"Approving LGTM on PR 42 after visual inspection\"\n  .\u002F.review-approved created\n\n# 3. Agent retries the action; this time it succeeds\n$ agent: gh pr review 42 --approve --body \"LGTM\"\n  [receipt appended to .\u002Freview-receipts\u002Freceipts.jsonl, decision=allow]\n\n# 4. Human closes the window\n$ rm .\u002F.review-approved\n```\n\nThe allowed call has a signed receipt that anyone with the public key can\nverify offline. The denied attempt has no receipt, and the approval log is\nnot signed, so keep both in mind when you show the trail to an auditor.\n\n## Composing with protect-mcp\n\nIf both plugins are installed, each plugin's `hooks\u002Fhooks.json` registers its\nown PreToolUse hook, and Claude Code runs both on every tool call:\n\n```json\n{ \"type\": \"command\", \"command\": \"\\\"${CLAUDE_PLUGIN_ROOT}\\\"\u002Fhooks\u002Fevaluate.sh\" }\n```\n\nEach `evaluate.sh` reads `tool_name` and `tool_input` from the hook payload on\nstdin (Claude Code sets no `TOOL_NAME` variable) and evaluates its own policy:\n`.\u002Fprotect.cedar` for protect-mcp and `.\u002Freview-governance.cedar` here.\n\nBoth hooks must pass for the tool call to proceed. Cedar deny in either\npolicy blocks it.\n\n## Standards\n\n- **Ed25519** — RFC 8032 (digital signatures)\n- **JCS** — RFC 8785 (deterministic JSON canonicalization)\n- **Cedar** — AWS's open authorization policy language\n- **IETF draft** — [draft-farley-acta-signed-receipts](https:\u002F\u002Fdatatracker.ietf.org\u002Fdoc\u002Fdraft-farley-acta-signed-receipts\u002F)\n",{"name":5,"description":7},"main","f846d9a97082ff30d613b55a085cef425d822473","plugins\u002Freview-agent-governance\u002Fskills\u002Freview-agent-setup","https:\u002F\u002Fgithub.com\u002Fshotcowboystyle\u002Fag-skills","https:\u002F\u002Fgithub.com\u002Fshotcowboystyle","https:\u002F\u002Fshotcowboystyle.github.io","Building smarter with AI, automation & modern tools. Exploring how far one person can go without a big team using today’s AI.","2026-11-10 04:58:27",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add shotcowboystyle\u002Fag-skills --skill review-agent-setup",{"tool":38,"command":39},"bunx","bunx skills add shotcowboystyle\u002Fag-skills --skill review-agent-setup",{"tool":41,"command":42},"pnpm","pnpm dlx skills add shotcowboystyle\u002Fag-skills --skill review-agent-setup","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fshotcowboystyle\u002Fag-skills\u002Fblob\u002Fmain\u002Fplugins\u002Freview-agent-governance\u002Fskills\u002Freview-agent-setup\u002FSKILL.md",true,[],[47,53,60,67,74,81],{"skillNo":48,"name":49,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M7F1YRY5S5QRJ1V0BJS130QX","no-description","https:\u002F\u002Fgithub.com\u002Fshotcowboystyle\u002Fag-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fno-description\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M7F1Y9BV2FJNQJ3NB32JS5CM","good-skill","Fixture skill that satisfies every rule.","https:\u002F\u002Fgithub.com\u002Fshotcowboystyle\u002Fag-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fgood-skill\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M7F1XXBQ3NQRXZSS0KAGAR9V","broken-link","Links to a reference file that does not exist.","https:\u002F\u002Fgithub.com\u002Fshotcowboystyle\u002Fag-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fbroken-link\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M7F1XBX4R46NR9W0HXQ7XCDX","Bad_Name","Name violates the kebab-case rule and does not match the directory.","https:\u002F\u002Fgithub.com\u002Fshotcowboystyle\u002Fag-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fbad-name\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M7F1WNAF8WNEQ7S5HKT5VK0D","skill-authoring","Write, review, and validate portable agent skills in SKILL.md format. Use when creating a new skill, editing an existing one, or asked whether a skill follows the Agent Skills spec.","https:\u002F\u002Fgithub.com\u002Fshotcowboystyle\u002Fag-skills\u002Fblob\u002Fmain\u002Fskills\u002Fskill-authoring\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M7F1W30TC0XFNSF7YBRGETK2","weigh-bags","Record scale readings for each bag — empty (tare), outbound, current, final — and reconcile them against the itemised inventory to find unaccounted weight. Use when the user gives a bag weight, says how heavy something is, weighs their luggage, or asks how close to the limit they are. Appends to bags.yaml and runs scripts\u002Freconcile.py.","https:\u002F\u002Fgithub.com\u002Fshotcowboystyle\u002Fag-skills\u002Fblob\u002Fmain\u002Fplugins\u002Ftravel-packing-assistant\u002Fskills\u002Fweigh-bags\u002FSKILL.md",[],[]]