Skip to content
Back to search

agent-reviews

MarketplaceWorks with Claude CodeGitHub

Token efficient, automated Github review workflows and bug fixing.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/pbakaus/agent-reviews with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/pbakaus/agent-reviews'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/pbakaus--agent-reviews/badge.svg)](/listings/pbakaus--agent-reviews)

HTML

<a href="/listings/pbakaus--agent-reviews"><img src="/api/listings/pbakaus--agent-reviews/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

CLI (npm)

npm install -g agent-reviews

Agent Skills

Three skills are available, each as a slash command (no npm install required):

| Skill | What it resolves | |-------|-----------------| | `resolve-reviews` | All comments (human + bot) | | `resolve-agent-reviews` | Bot comments only (Copilot, Cursor, etc.) | | `resolve-human-reviews` | Human comments only |

Works with any agent that supports [Agent Skills](https://agentskills.io) (Claude Code, Cursor, Codex, etc.):

npx skills add pbakaus/agent-reviews@resolve-agent-reviews

Replace `resolve-agent-reviews` with whichever skill you want. Skills use `npx agent-reviews` at runtime, so the CLI is fetched automatically.

> You can also use both: install the CLI globally for direct terminal use, and a skill for the agent workflow.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 26 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "name": "agent-reviews",
  3. "owner": {
  4. "name": "Paul Bakaus"
  5. },
  6. "metadata": {
  7. "description": "CLI and Claude Code skills for managing GitHub PR review comments"
  8. },
  9. "plugins": [
  10. {
  11. "name": "resolve-reviews",
  12. "source": "./",
  13. "description": "Resolve all PR review comments (human and bot) on the current PR"
  14. },
  15. {
  16. "name": "resolve-agent-reviews",
  17. "source": "./",
  18. "description": "Resolve PR review bot findings (Copilot, Cursor Bugbot, etc.) on the current PR"
  19. },
  20. {
  21. "name": "resolve-human-reviews",
  22. "source": "./",
  23. "description": "Resolve human PR review comments on the current PR"
  24. }
  25. ]
  26. }
README.md · README · 279 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # agent-reviews
  2. Manage GitHub PR review comments from the terminal and from AI coding agents.
  3. PR review bots (Copilot, Cursor Bugbot, CodeRabbit, etc.) leave inline comments on your pull requests. agent-reviews gives you a CLI to list, filter, reply to, and watch those comments, plus agent skills that automate the entire triage-fix-reply loop.
  4. ## Why
  5. **`gh` CLI is fragile for review comments.** Agents frequently get the syntax wrong, fail to paginate, and can't reliably detect whether a comment has been replied to. agent-reviews provides a single, purpose-built interface that handles all of this correctly.
  6. **Bot reviews create a doom loop.** You fix one round of findings, push, and new comments appear. Fix those, push again, more comments. This cycle can eat hours. The included skills solve this with an integrated watcher that keeps fixing and replying until the bots go quiet.
  7. **Works in cloud environments.** Most solutions rely on local tooling that isn't available in cloud or remote agent environments. agent-reviews works everywhere, so you can kick off a session, let the agent resolve all findings autonomously, and come back to a clean PR.
  8. ## Install
  9. ### CLI (npm)
  10. ```bash
  11. npm install -g agent-reviews
  12. ```
  13. ### Agent Skills
  14. Three skills are available, each as a slash command (no npm install required):
  15. | Skill | What it resolves |
  16. |-------|-----------------|
  17. | `resolve-reviews` | All comments (human + bot) |
  18. | `resolve-agent-reviews` | Bot comments only (Copilot, Cursor, etc.) |
  19. | `resolve-human-reviews` | Human comments only |
  20. Works with any agent that supports [Agent Skills](https://agentskills.io) (Claude Code, Cursor, Codex, etc.):
  21. ```bash
  22. npx skills add pbakaus/agent-reviews@resolve-agent-reviews
  23. ```
  24. Replace `resolve-agent-reviews` with whichever skill you want. Skills use `npx agent-reviews` at runtime, so the CLI is fetched automatically.
  25. > You can also use both: install the CLI globally for direct terminal use, and a skill for the agent workflow.
  26. ## Authentication
  27. The simplest method is the **GitHub CLI**. If you're logged in with `gh auth login`, agent-reviews picks up the token automatically. No configuration needed.
  28. For cloud/remote environments or HTTPS proxy setups, set `GITHUB_TOKEN` or `GH_TOKEN` directly. agent-reviews includes [undici](https://github.com/nodejs/undici) `ProxyAgent` support and will route requests through `HTTPS_PROXY` automatically when set.
  29. **Resolution order** (first match wins):
  30. 1. `GITHUB_TOKEN` environment variable
  31. 2. `GH_TOKEN` environment variable
  32. 3. `.env.local` in the repo root
  33. 4. `gh auth token` (GitHub CLI)
  34. ### Fork pull requests
  35. Branch discovery keeps the head repository separate from the PR's base repository. It uses the branch push remote, `remote.pushDefault`, tracking remote, or `origin` (in that order), and looks for the current branch in configured GitHub repositories and the fork's parent/source repositories. The selected remote's push URL determines the head repository, including when its fetch URL points upstream. An upstream remote is not required. A single GitHub remote with another name also works.
  36. `GH_REPO=organisation/project` restricts the target repository without changing the head owner. Use `GH_REPO=organisation/project agent-reviews --pr 123` to select a PR explicitly, including from detached HEAD. Malformed `GH_REPO` values fail explicitly. If multiple open PRs match, or a matching PR has a deleted/inaccessible head repository that cannot be verified, the CLI asks for an explicit target instead of choosing one. Discovery uses the GitHub API and works with token authentication without requiring `gh`.
  37. ### Custom API host
  38. Set `GITHUB_API_URL` to point agent-reviews at a GitHub Enterprise host or any API-compatible server (useful for testing, recording, or routing through a local mediator). Defaults to `https://api.github.com`.
  39. ```bash
  40. # GitHub Enterprise Server
  41. GITHUB_API_URL=https://github.example.com/api/v3 agent-reviews
  42. # Local API-compatible server
  43. GITHUB_API_URL=http://127.0.0.1:8080 agent-reviews
  44. ```
  45. GraphQL endpoint resolution: agent-reviews uses `${GITHUB_API_URL}/graphql` by default. For GitHub Enterprise Server, where REST lives under `/api/v3` and GraphQL under `/api/graphql` on the same origin, the trailing `/api/v3` is rewritten to `/api/graphql` automatically. Set `GITHUB_GRAPHQL_URL` directly if you need full control over the GraphQL endpoint.
  46. ## CLI Usage
  47. ```bash
  48. # List all review comments on the current branch's PR
  49. agent-reviews
  50. # Only unresolved comments
  51. agent-reviews --unresolved
  52. # Only unanswered bot comments
  53. agent-reviews --unanswered --bots-only
  54. # Full detail for a specific comment (diff hunk + replies)
  55. agent-reviews --detail 12345678
  56. # Reply to a comment
  57. agent-reviews --reply 12345678 "Fixed in abc1234"
  58. agent-reviews --reply 12345678 --body-file body.md
  59. # JSON output for scripting / AI agents
  60. agent-reviews --json
  61. # Watch for new comments (polls every 30s, exits after 10 min idle)
  62. agent-reviews --watch --bots-only
  63. # After a push: wait for the head commit's checks, then print failures and unanswered bot comments
  64. agent-reviews --wait-checks --bots-only --expect "Greptile Review"
  65. # Target a specific PR (otherwise auto-detects from branch)
  66. agent-reviews --pr 42
  67. ```
  68. Unknown options, unexpected arguments, and invalid numeric values fail before authentication or network requests. To pass an inline reply starting with a dash, put options first and use `agent-reviews --reply 12345 -- "- Fixed the issue"`. Everything after `--` is literal text.
  69. ### Options
  70. | Flag | Short | Description |
  71. |------|-------|-------------|
  72. | `--unresolved` | `-u` | Only unresolved/pending comments |
  73. | `--unanswered` | `-a` | Only comments without any replies |
  74. | `--reply <id> "msg"` | `-r` | Reply to a comment |
  75. | `--body-file <path>` | | Read reply text from a UTF-8 file instead of a positional message |
  76. | `--resolve` | | Resolve the review thread after replying (use with `--reply`) |
  77. | `--detail <id>` | `-d` | Full detail for a comment |
  78. | `--pr <number>` | `-p` | Target a specific PR |
  79. | `--json` | `-j` | JSON output |
  80. | `--bots-only` | `-b` | Only bot comments |
  81. | `--humans-only` | `-H` | Only human comments |
  82. | `--ignore-author <login>` | | Exclude an author from listing and watching; repeat for multiple authors |
  83. | `--expanded` | `-e` | Show full detail for each listed comment |
  84. | `--watch` | `-w` | Poll for new comments |
  85. | `--wait-checks` | | Wait for the head commit's checks, then report failures and unanswered comments |
  86. | `--expect <name>` | | With `--wait-checks`: a check that must report before the wait ends; repeatable |
  87. | `--interval <sec>` | `-i` | Poll interval in seconds (default: 30, or 15 with `--wait-checks`) |
  88. | `--timeout <sec>` | | Inactivity timeout in seconds (default: 600); with `--wait-checks`, the total wait (default: 1800) |
  89. To hide workflow status comments, use `agent-reviews --ignore-author 'github-actions[bot]'`. Repeat the flag for multiple authors, including with `--watch`. Matching is case-insensitive and uses the full GitHub login, including `[bot]` when present. Excluded authors' top-level findings are hidden, but their replies remain visible as context in other authors' threads. Explicit `--detail` lookups and replies are unaffected. No authors are excluded by default.
  90. ## Agent Skills
  91. The skills automate the full PR review resolution workflow:
  92. 1. Fetch unanswered comments (all, bot-only, or human-only depending on skill)
  93. 2. Evaluate each finding (true positive, false positive, actionable, etc.)
  94. 3. Fix real issues and run lint/type-check
  95. 4. Dismiss false positives with an explanation
  96. 5. Reply to every comment with the outcome
  97. 6. Wait for the head commit's checks (`--wait-checks`) and repeat until a pass is clean
  98. 7. Report a summary of all actions taken
  99. ### Skill behavior
  100. - **True positives / actionable feedback** get fixed and replied with `Fixed in {commit}`
  101. - **False positives** get replied with `Won't fix: {reason}`
  102. - **Uncertain findings** prompt the user for guidance
  103. - All fixes are batched into a single commit before polling begins
  104. - The bot and combined skills loop until every check on the head commit has passed and no comment waits; the human-review skill watches until reviewers go quiet
  105. ## How It Works
  106. ### Comment types
  107. agent-reviews fetches three types of GitHub PR comments:
  108. | Type | Label | Description |
  109. |------|-------|-------------|
  110. | Review comment | `CODE` | Inline comment attached to a specific line |
  111. | Issue comment | `COMMENT` | General PR-level comment |
  112. | Review | `REVIEW` | Review summary (approved, changes requested) |
  113. ### Meta-comment filtering
  114. Bot review bodies (`REVIEW` type) are always filtered out since actionable findings come through as inline comments. Additionally, these bot issue comments are filtered:
  115. | Bot | What's filtered |
  116. |-----|----------------|
  117. | Vercel | Deployment status (`[vc]:...`) |
  118. | Supabase | Branch status (`[supa]:...`) |
  119. | Cursor Bugbot | Review summary ("Cursor Bugbot has reviewed your changes...") |
  120. | Copilot | PR review summary ("Pull request overview") |
  121. | CodeRabbit | Walkthrough, summary, and "review skipped" comments |
  122. | Sourcery | Reviewer's guide and PR summary |
  123. | Codacy | Analysis summary and coverage summary |
  124. | SonarCloud | Quality Gate pass/fail summary |
  125. | Gemini Code Assist | "Summary of Changes" issue comment |
  126. ### Reply status
  127. Each comment displays its reply status:
  128. | Status | Meaning |
  129. |--------|---------|
  130. | `no reply` | No one has replied |
  131. | `replied` | A human has replied |
  132. | `bot replied` | Only bots have replied |
  133. ### Waiting for checks
  134. `--wait-checks` polls the PR's head commit until every check run and commit status has finished, then prints one line with the outcome, the failed checks with links, any expected check that never reported, and the unanswered comments (with the usual filters). It follows new pushes to the PR, stops at once when the PR conflicts with its base branch (no CI runs then), and waits 30 seconds after the last check finishes for reviewers that register late. Name a reviewer with `--expect` when it registers much later, for example only after a preview deploys; expected checks are not waited for on drafts.
  135. | Exit code | Meaning |
  136. |---|---|
  137. | 0 | Every check passed and no comment waits |
  138. | 1 | A check failed, an expected check never reported, or an inline review comment waits |
  139. | 2 | The PR conflicts with its base branch |
  140. | 3 | Timed out with checks still running |
  141. | 4 | No check registered on the commit |
  142. Summary comments from review bots (Greptile, Copilot, CodeRabbit and others) are not counted as findings. PR-level comments (issue comments, review bodies) are listed but never change the exit code, since they have no thread to reply in and would otherwise hold every run at 1.
  143. ### Watch mode
  144. Polls the GitHub API at a configurable interval and reports new comments as they appear. Outputs both formatted text and JSON for AI agent consumption. Exits automatically after a configurable inactivity timeout (default: 10 minutes).
  145. Reply files preserve Markdown and newlines exactly. For multiline or complex replies, write the text to a file and pass `--body-file body.md` to avoid shell quoting and command substitution. Both absolute and relative paths are supported; relative paths are resolved from the current directory. Prefer uniquely named temporary files outside the repository so later commits cannot accidentally include reply text. Use either a positional message or `--body-file`, not both; empty or whitespace-only replies are rejected. The flag works with `--resolve` and `--json`.
  146. ## Changelog
  147. ### 1.3.0 (October 16, 2026)
  148. - The bot and combined skills stop after a second round of fixes when new findings keep landing on lines the earlier fixes changed, and hand the rest to the user instead of patching on.
  149. - `--wait-checks` rides out network drops of up to five minutes instead of five failed requests, and retries the PR lookup and the comment fetch through short drops.
  150. ### 1.2.0 (October 15, 2026)
  151. - Wait for a PR's checks with `--wait-checks`: follows new pushes, stops on a conflict with the base branch, waits for late reviewers named with `--expect`, and prints only failed checks and unanswered comments, with an exit code for each outcome.
  152. - The bot and combined skills wait with `--wait-checks` instead of a quiet window.
  153. - Treat Greptile's PR summary as a summary, not a finding.
  154. ### 1.1.0 (September 10, 2026)
  155. - Read exact UTF-8 reply text with `--body-file`, without shell command substitution.
  156. - Reject unknown flags and invalid arguments before posting; use `--` for replies beginning with a dash.
  157. - Discover fork PRs across upstream repositories, separate push URLs, and GitHub Enterprise hosts.
  158. - Exclude selected authors from listing and watching with repeatable `--ignore-author`.
  159. - Reduce duplicate replies with fresh reply checks and watcher-startup reconciliation in all three skills.
  160. - Update development dependencies to Vitest 5 and refresh website dependencies and audit fixes.
  161. Development tests now require Node.js 22.12+, 24, or 26+ as supported by Vitest 5. CLI runtime support remains Node.js 18+.
  162. ### 1.0.2
  163. - GitHub Enterprise Server support via `GITHUB_API_URL` env var (also works for local API-compatible servers used in testing). REST and GraphQL endpoints both honor it, with GHES `/api/v3` automatically rewritten to `/api/graphql`. Optional `GITHUB_GRAPHQL_URL` for unusual setups.
  164. - Gemini Code Assist meta-comment filtering: the bot's `## Summary of Changes` issue comment is now dropped, while inline severity-badged findings are preserved.
  165. ### 1.0.0
  166. **Three skills instead of one.** The single `agent-reviews` skill has been split into three, each tailored for different workflows:
  167. - `resolve-reviews` resolves all comments (human + bot)
  168. - `resolve-agent-reviews` resolves bot comments only
  169. - `resolve-human-reviews` resolves human comments only
  170. **Thread resolution.** The new `--resolve` flag marks GitHub review threads as resolved after replying. Uses the GraphQL `resolveReviewThread` mutation. Works with `--reply` in any argument order.
  171. **Expanded bot support.** Added detection and meta-comment filtering for CodeRabbit, Sourcery, Codacy, SonarCloud/SonarQube Cloud, and Copilot PR reviewer, in addition to the existing Cursor Bugbot, Vercel, and Supabase filters.
  172. **Agent-harness universal.** Skills now work with any agent that supports [Agent Skills](https://agentskills.io) (Claude Code, Cursor, Codex, etc.), not just Claude Code.
  173. **Watch mode improvements.** The watcher now exits immediately when new comments are found (with a 5s grace period for batch posts), designed for loop-based workflows where the agent processes comments and restarts the watcher.
  174. **New CLI options:**
  175. - `--resolve` resolves the review thread after replying (use with `--reply`)
  176. - `--expanded` / `-e` shows full detail (body, diff hunk, replies) for each comment in list mode
  177. **Bug fixes:**
  178. - `--json --resolve` no longer emits plain-text status messages to stdout
  179. **Cloud and proxy support:**
  180. - `GH_TOKEN` environment variable support (in addition to `GITHUB_TOKEN`)
  181. - `GH_REPO` environment variable for targeting repos in detached environments
  182. - Curl-based HTTP fallback for environments without native fetch/undici
  183. - Curl requests include timeouts (10s connect, 60s max)
  184. **Smarter filtering.** Bot review bodies (summaries listing inline findings) are now automatically excluded, since actionable findings always come through as inline comments. Reply comments posted by agent-reviews itself (`> Re: comment ...`) are also filtered to avoid noise.
  185. **Simplified architecture.** Skills now invoke `npx agent-reviews` at runtime instead of bundling their own scripts, reducing the package from ~4000 lines of duplicated code to a single CLI entry point. Skills no longer run redundant startup commands (version check, branch detection, PR lookup), relying on the CLI's own error handling instead.
  186. ## License
  187. MIT

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.