Borrowing it
Nothing to install: this file belongs to r3bl-org/r3bl-open-core. 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/r3bl-org/r3bl-open-core/main/.agents/skills/review-pr/SKILL.mdgit clone --depth 1 https://github.com/r3bl-org/r3bl-open-coreWrote 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/skills/r3bl-org/r3bl-open-core/review-pr)<a href="https://agentmods.dev/skills/r3bl-org/r3bl-open-core/review-pr"><img src="https://agentmods.dev/badge/skills/r3bl-org/r3bl-open-core/review-pr/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/skills/r3bl-org/r3bl-open-core/review-pr"><img src="https://agentmods.dev/badge/skills/r3bl-org/r3bl-open-core/review-pr.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 27 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00014 | $0.00856 |
| Opus 5 | $0.00007 | $0.00428 |
| Sonnet 5 | $0.00003 | $0.00171 |
| Haiku 4.5 | $0.00001 | $0.00086 |
Grade A, and why
review-pr 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 12d 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR Review and Integration Workflow
Use this skill when the user runs the /review-pr <number> slash command or uses any of
these natural language triggers:
- "lets review pr "
- "lets work on pr "
- "lets take a look at pr "
Workflow
When triggered, you MUST follow these exact steps:
- Information Gathering
- Run
gh pr view <number>to read the PR description. - Run
gh pr diff <number>to see the exact code changes and files touched.
- Run
- Task File Generation
- Create a new markdown task file at
task/pr-<number>-fix.md. - The file MUST follow the template structure below, organizing the PR into distinct, actionable chunks (phases/headings).
- Create a new markdown task file at
- Format & Present
- Automatically run
prettier --write task/pr-<number>-fix.mdto format the new file. - Run
antigravity-ide task/pr-<number>-fix.mdto open the file for the user. - Ask the user to manually review and explicitly approve the plan before any implementation begins.
- Automatically run
Task File Template
Canonical Example: See task/done/pr-455-fix.md for a complete, canonical example of a well-written PR review task file.
When creating task/pr-<number>-fix.md, structure it exactly like this:
_Task: PR <number> Integration (<short description>)_
# User Story & Context
## Problem
[Describe the user-facing problem that this PR is attempting to solve. What triggers the
bug?]
## Root Cause
[Describe the underlying technical root cause. Why is this happening in the codebase?]
## Expected Behavior
[Describe what the expected behavior should be after the fix.]
# Overview
[Brief summary of the PR, author, and what problem it solves.]
# Implementation Plan
We will process each of the action items iteratively using the following loop:
1. **Implementation:** Write the specific code changes for the current heading.
2. **Local Testing:** Run `./check.fish --check` and, where applicable, test
functionality.
3. **Mandatory Manual Review:** You (the user) will manually review the specifically
touched files before the heading is marked as checked `[x]`.
_(Once all headings are successfully implemented and checked off, we will proceed to final
verification and cleanup.)_
## Phase 1: [Name of first distinct fix/feature]
[Brief context of the problem and the specific fix.]
- _Context:_ [Why this change is needed.]
- _The Fix:_ [What the code actually does.]
- _File(s) Touched:_ [List the files.]
## Phase 2: [Name of next fix/feature...]
...
## Final Verification & Cleanup
- [ ] Verify full test suite coverage using `./check.fish --full`.
- [ ] When ready to merge, invoke the `/merge-pr` slash command to push the changes,
optionally update the author, and cleanly merge to `main`.
- [ ] Update the current meta-task (e.g. `task/prepare-vX.Y.Z-meta-task.md`) to check off
PR #<number>.
- [ ] **Mandatory manual review:** Verify every file modified in this task for correct
implementation and ensure no regressions.
- [ ] `path/to/modified_file_1.rs`
- [ ] `path/to/modified_file_2.rs`
- [ ] `task/prepare-vX.Y.Z-meta-task.md`
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.
- 12d ago First seen · 106 lines · 14 tokens per session scan A e4eedcc96a6c
review-pr is a skill published in the GitHub repository r3bl-org/r3bl-open-core (483 stars, last pushed today), licensed Apache-2.0. It adds 14 tokens to every session and 856 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 2026-08-30.
Other skills, from other repositories
cw-slice
Use before writing code for any Codewhale feature, upgrade, or refactor: find the existing owner of the behavior, bound the change to one reviewable slice, and fix the evidence bar before you start.
gh-compile-issues
Triage N GitHub issues into a coverage matrix: fetch each, check current code, classify already-done/quick-fix/design/defer with cited evidence.
gh-find-prs
Survey open Codewhale PRs and triage each for mergeability and disposition against the real landing branch.
review
Diff-scoped correctness review that reads the codebase around the change — callers, contracts, and invariants — and returns line-anchored findings ranked by severity with confidence, then a merge-risk verdict. Use for reviewing a PR, diff, or named change set. Not for style review, approvals-as-rubber-stamp, or…
simplify
Improve clarity and reduce needless complexity after behavior is understood; preserve behavior and keep cleanup separate from correctness fixes.
cmd-review
Review the current diff or selected files (usage: /review [--last-diff|--target |--file |files...] [--style ]).