{ "markdown": "

\n \n \n \"Open\n \n

\n\n

\n AI code review that lands directly on your PR.
\n Open source. Self-hosted.
\n Works with \"\" GitHub · \"\" GitLab · \"\" Bitbucket
\n /open-pr:review · /open-pr:fix\n

\n\n

\n \"Release\"\n \"License:\n \"Listed\n \"GitHub\"\n \"GitLab\"\n \"Bitbucket\"\n

\n\n

\n \"Claude\n \"Cursor\"\n \"Codex\"\n \"Gemini\n \"Antigravity\"\n

\n\n

\n Tiếng Việt · English · 日本語 · 简体中文\n

\n\nAI coding made PRs faster. But review didn't get faster.\n\n**`open-pr` runs that first review round for you — on the PR, not on your laptop.** Anyone who opens the PR sees the same feedback.\n\n

\n \"Overview,\n

\n\nOne run produces three parts that belong together: an **overview**, **line comments** (with suggested changes), and a **reply** after `/open-pr:fix` has pushed. — [See the demo](./docs/demo.md)\n\n- 🔍 **Exactly 1 review per run** — one review posted, not a stream of bot comments\n- 🧠 **Learns your repo** — README / CLAUDE.md / AGENTS.md / docs / wiki; team rules beat generic rules\n- 💬 **Remembers what the team said** — a correction on one PR carries into the next run\n- 🔧 **`/open-pr:fix` is disciplined** — exactly **1** commit, no force-push, a reply on every thread\n- 🔓 **Open source, no `open-pr` service** — MIT-licensed, no `open-pr` server or bot account; it runs in the agent CLI you already have\n\n## Listed in Awesome AI Plugins\n\n`open-pr` is listed in [Awesome AI Plugins](https://github.com/hashgraph-online/awesome-ai-plugins#development--workflow) — the cross-platform catalog curated by Hashgraph Online — under *Community Plugins → Development & Workflow*, and it is ingested by the [HOL Plugin Registry](https://hol.org/registry/plugins), which scans every listed project and publishes a trust score.\n\nThe listing is gated on that scan: **score ≥ 80, with no high or critical finding**. The catalog runs the scanner itself against this repository's default branch, so the result is theirs rather than ours. A scan is a trust signal, not a safety guarantee.\n\n## Install\n\n**1. A vendor CLI, logged in.** The plugin carries no credential of its own — it reads the PR and posts the review through *your* account:\n\n```bash\n# GitHub\nbrew install gh # or https://cli.github.com/\ngh auth login # GitHub.com → HTTPS → Login with a web browser\ngh auth status # must say \"Logged in to github.com as \"\n```\n\nGitLab: `brew install glab && glab auth login --hostname gitlab.com`. Bitbucket ships no CLI — it reads `BITBUCKET_EMAIL` + `BITBUCKET_API_TOKEN` from the environment. Minimum permissions and how to check them: [Getting a token per vendor](./docs/credentials.md).\n\n**2. The plugin.** [Claude Code](https://claude.ai/code):\n\n```bash\n/plugin marketplace add TOMOSIA-VIETNAM/open-pr\n/plugin install open-pr@open-pr\n```\n\n**Install for Cursor, Codex, Gemini CLI, Antigravity:**\n\n```bash\ncurl -fsSL https://raw.githubusercontent.com/TOMOSIA-VIETNAM/open-pr/main/install.sh | bash\n```\n\nFull guide: [Install](./docs/install.md) · [Getting a token per vendor](./docs/credentials.md).\n\nPR URL formats: GitHub `.../pull/` · GitLab `.../-/merge_requests/` (self-hosted included) · Bitbucket Cloud `.../pull-requests/`.\n\n## Why review is the bottleneck\n\n

\n \n \n \"Before\n \n

\n\nIn the age of AI coding, PRs ship far faster than they get reviewed. The bottleneck is no longer coding — it's **review**. Reviewers have to check the project's conventions / security / performance *and* cover business logic — and that doesn't scale well as PR volume grows.\n\nA local review is hard to trust. Anyone can say *\"I already reviewed it\"*. So `open-pr` moves that step to **remote** for transparency — comments sit on the PR, and anyone who opens it can see them.\n\n> [!NOTE]\n> **Review rounds** (a suggestion for the team):\n> 1. **Round 1** — Dev runs AI review on the PR themselves. No review comments yet → reviewer **sends it back**, without touching it.\n> 2. **Round 2** — Reviewer runs it again (AI). Clean → **LGTM**.\n> 3. **Round 3** — Reviewer reviews the domain part.\n\n> [!IMPORTANT]\n> AI lightens the process load, but **final responsibility is still yours**.\n\n## How it differs from a generic review skill\n\nMany review skills are just a `SKILL.md` description. Each run comes out differently — different wording, different strictness, easy to drift from the project's conventions.\n\n| Common with a generic skill | With `open-pr` |\n| --- | --- |\n| Advice stays at generic rules, off the project | Reads README / CLAUDE.md / AGENTS.md / docs / wiki; **team rules beat** generic rules |\n| You remind it once, next time it slips again | Mentions in chat → asks to write into the repo's memory → next run applies it |\n| Told to fix → fixes per the comment — even a wrong comment → correct code becomes wrong | `/open-pr:fix` weighs whether a comment is sound; if not → **reply + evidence**, no code change |\n| Fixes arrive as commit spam, amends, force-pushes, no replies | Exactly **1 commit** per `fix`, no history rewriting, a reply per comment after push |\n\n> [!TIP]\n> The part worth keeping: no matter when you run it, the procedure is the same — bootstrap conventions, pick the output language from the repo, then remember what the team has reminded. Not one AI voice today and another tomorrow.\n\n## Review flow\n\n```mermaid\nflowchart LR\n A[New PR] --> B[\"Round 1 · /open-pr:review\"]\n B --> C{Review on remote?}\n C -- not yet --> D[Reviewer sends it back]\n C -- yes --> E[Dev fix / /open-pr:fix]\n E --> F[\"Round 2 · review again\"]\n F --> G{Clean?}\n G -- yes --> H[LGTM]\n G -- not yet --> E\n H --> I[Round 3 · human domain review]\n```\n\nDetails on re-review, worktrees, and the guard before `fix`: [Re-review / fix flow](./docs/how-it-works.md).\n\n## Commands\n\n| Command | What it does |\n| --- | --- |\n| `/open-pr:review ` | Posts exactly **1** review. No code edits, no close, no merge. First run in a repo also sets it up |\n| `/open-pr:fix ` | Reads findings → weighs right/wrong → fixes → **1** commit → replies. 🔵 / 📝 always ask first |\n| `/open-pr:upgrade` | Brings local config up to the current schema — summarises, then asks; nothing written until you agree |\n| `/open-pr:clean` | Removes worktrees that `review` checked out (asks first). Memory / settings untouched |\n| `/open-pr:feedback` | Reports a problem with **this plugin** on its issue tracker — stripped of anything identifying your repo, and shown to you before it is posted |\n\n> [!WARNING]\n> `fix` edits **real code** in the repo (or the review worktree). Run it only when you deliberately want it to handle the comments.\n\nFull configuration: [Configuration](./docs/configuration.md).\n\n## What it reviews\n\n1. **Bugs & logic**\n2. **Security**\n3. **Performance**\n4. **Code quality**\n5. **Maintainability & readability**\n6. **Framework / language-specific** — from that stack's template\n\nCriteria in detail and priority when they conflict: [What it reviews](./docs/review-criteria.md).\n\n## Prompt token chart\n\nMean tokens per run — covering both *happy-case* and *bad-case*:\n\n![Mean tokens per run, by command / release](./token-history.svg)\n\n---\n\nContributing? [CONTRIBUTING.md](./CONTRIBUTING.md).\n\nEnjoy reviewing 🥰\n\n

\n \n \n \"\"\n \n

\n\n

\n Logo files: docs/logo.md\n

\n", "bytes": 10470, "sha": "15a6c26e301c53bbcd6f09ec882df2dc616523647757a69aa2b863d11a5f05bb", "repo_slug": "tomosia-vietnam/open-pr", "fonte": "repo", "truncated": false, "api": "https://agentalog.com/api/listings/plg_tomosia_vietnam_open_pr_422fbd13/readme" }