- Home
- Skills
- Code Quality & Review
- ELYRAON PR Risk Reviewer
Works with the AI Agents you already use
ELYRAON PR Risk Reviewer
Review a patch for concrete regression risk before it lands.
$7.99
About this skill
A generic diff summary tells you what changed. ELYRAON PR Risk Reviewer looks for concrete risk signals in a proposed change, so you know what to verify before it lands.
What it does
- Diff validation checks that the input is a supported unified text diff and rejects binary content and combined patches instead of guessing.
- Risk identification flags changes to security-sensitive paths, CI configuration, and dependency manifests.
- Line-level evidence points to the specific hunks and relative file paths behind each finding, so you can check them yourself.
- Static analysis runs locally through a bundled Python helper on the diff file you provide. It does not execute your code.
- JSON reporting produces structured output with risk codes and findings, which your agent turns into a review with recommended follow-up verification.
How it works
- Point the skill at a unified diff file.
- A bundled Python 3 script parses the diff and inventories the changed hunks.
- Your agent reads the JSON report and presents findings with their evidence.
- You decide what to verify, test, or fix before merging.
Requirements
Python 3 and a unified diff file of up to 1MB. No repository access or external services are needed.
What it is not
It is a static review aid. It does not run tests, create or merge pull requests, modify your code, or prove a patch is safe. See Known Limitations.
Analyzes unified text diffs only, up to 1MB per file. It does not support binary files, renames, or combined patches. Analysis is static: it does not run tests or your code and does not check whether changed code is reachable at runtime. It sees only the diff you provide, with no repository-wide context, so incomplete diffs reduce confidence. Findings are evidence for human review, not a guarantee: the absence of findings does not prove a patch is safe, and it is not a complete security review. It does not modify your code or create or merge pull requests.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 month ago
Requires
Permissions
Terminal/Shell: runs the bundled Python helper locally on the diff file you provide. Read Files: reads that diff file. No network access, no file writes, no environment variables.