Borrowing it
Nothing to install: this file belongs to TOMOSIA-VIETNAM/open-pr. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/TOMOSIA-VIETNAM/open-pr/main/CLAUDE.mdgit clone --depth 1 https://github.com/TOMOSIA-VIETNAM/open-prWrote 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/instructions/tomosia-vietnam/open-pr/claude-md)<a href="https://agentmods.dev/instructions/tomosia-vietnam/open-pr/claude-md"><img src="https://agentmods.dev/badge/instructions/tomosia-vietnam/open-pr/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/tomosia-vietnam/open-pr/claude-md"><img src="https://agentmods.dev/badge/instructions/tomosia-vietnam/open-pr/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.02680 | $0.02680 |
| Opus 5.5 | $0.01072 | $0.01072 |
| Sonnet 5.5 | $0.00536 | $0.00536 |
| Haiku 4.5 | $0.00268 | $0.00268 |
Grade A, and why
open-pr CLAUDE.md scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| `/open-pr:review <PR_URL>` | review PR/MR, learn repo conventions, post one review via vendor CLI (`gh`/`glab`) or REST/`curl` | How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
How an agent works in this repo. Orientation + durable invariants — not a ritual to game.
Plugin
Claude Code plugin open-pr. Markdown + one JSON config + ONE POSIX-sh runtime (src/bin/open-pr.sh) that owns every vendor/git mechanic, plus src/bin/open-pr-watch.sh, which owns watch state and each platform's session CLI and reaches vendors only through open-pr.sh. No build. Only src/ ships (/plugin install copies src/).
| command | does |
|---|---|
/open-pr:review <PR_URL> |
review PR/MR, learn repo conventions, post one review via vendor CLI (gh/glab) or REST/curl |
/open-pr:fix <PR_URL> |
read findings, edit real code at pwd, one commit, reply on the PR |
/open-pr:upgrade |
migrate this repo's config to latest schema_version; fetches llm-upgrades/ live |
/open-pr:clean |
remove review worktrees after confirm. Never touches memory or config |
/open-pr:watch [review|fix] |
watch the repos picked under pwd: 1 review session per /open-pr PR comment, 1 fix session per own PR's new findings on "Fix now" (claude --bg or a platform's headless CLI); relay questions/drafts |
/open-pr:menubar [close] |
macOS menu bar item for every watcher on this machine: reviews and fixes in progress, "Fix now", recent toasts, snooze |
/open-pr:feedback |
turn this chat into one issue on this plugin's tracker (user-approved, de-identified) |
Vendors: GitHub, GitLab, Bitbucket.
Layout
src/bin/ open-pr.sh — the deterministic runtime · open-pr-watch.sh — watch state + session runners · open-pr-toast.js — macOS toast · open-pr-menubar.js — macOS menu bar · open-pr-session-title.sh — names the session; code, never `Read` into context except by `cases/self-repair.md`
src/hooks/ hooks.json — Claude Code hooks, auto-loaded from the plugin root
src/commands/ entry points; only these have frontmatter
src/core/ shared procedure
src/setup/ bootstrap, doctor, template, lesson
src/cases/ gated branches — Read only when the condition matches
src/templates/ per-stack criteria, copied into the reviewed repo
src/reference/ schema + vendor contract — do not Read at run time
src/seeds/ copied verbatim into the reviewed repo — never Read back
llm-upgrades/ config migrations, fetched live, not packaged
adapters/ ROOT + tool-name map; the only agent-read file naming a non-Claude platform (open-pr-watch.sh, install*.sh, commands/*.toml name them too)
skills/ one shim per command (Cursor · Codex · Gemini · Antigravity)
commands/*.toml Gemini CLI entry format
install.sh non-Claude install: clone → install-local.sh
scripts/ check.sh · token_report.py · dup_scan.py · vendor_lint.py · cli_doc.py · hooks · install-local.sh
tests/ graph/ (one file per invariant family) · test_cli.py · budgets.json
duplication_allowlist.json · ordered_repeats.json
e2e/ real-run fixture; never in CI
.claude/skills/ dev skills (`e2e-loop`)
.github/workflows ci.yml on PRs and main · hol-plugin-scanner.yml read by the listing gate
backlogs/ historical, not ops
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.
- 4d ago Changed · +3 lines · +205 tokens per session c93ce581ce85
- 12d ago Changed · +2 lines · +208 tokens per session 487e147e55f3
- 1mo ago Changed · +1 lines · +55 tokens per session 3cbea4bd0461
- 1mo ago Changed · +4 lines · +202 tokens per session d90dbff2b60d
- 1mo ago Changed · +1 lines · +86 tokens per session 49b73b795f35
- 2mo ago First seen · 105 lines · 1,924 tokens per session scan A 09a5f161fe3b
open-pr CLAUDE.md is an instructions file published in the GitHub repository TOMOSIA-VIETNAM/open-pr (44 stars, last pushed 3d ago), licensed MIT. It adds 2,680 tokens to every session, about $0.0107 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.