Performs findings-first code reviews to identify bugs, validation gaps, and deployment risks in local code.

Install

mkdir -p .claude/skills/risk-review && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/19339" && unzip -o skill.zip -d .claude/skills/risk-review && rm skill.zip

Installs to .claude/skills/risk-review

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Repo-local findings-first review workflow for Zoolanding Lambda diffs, deploy candidates, and contract changes. Use when reviewing for bugs, regressions, rollout risk, weak validation, or missing tests.
202 chars✓ has a “when” trigger
Advanced

Key capabilities

  • Identify bugs and release risks in diffs
  • Assess deploy-readiness of lambda functions
  • Analyze request-response contract changes
  • Challenge unproven assumptions in code
  • Ground findings in concrete evidence

How it works

The workflow focuses on high-risk surfaces like validation and state transitions, requiring reviewers to provide evidence-based findings and suggest specific fixes.

Inputs & outputs

You give it
Code diffs or pull requests
You get back
Prioritized list of findings with severity levels

When to use risk-review

  • Review pull requests for potential bugs
  • Assess deploy-readiness of lambda functions
  • Identify missing tests or validation logic
  • Analyze changes to request-response contracts

About this skill

Risk Review

This repo-local version is for review-only work. Lead with findings, not summaries.

Review Targets

  • local diffs or pull requests
  • request or response contract changes
  • storage, lifecycle, or payload behavior changes
  • deploy-readiness and regression checks

Output Contract

Each finding should include:

  • severity
  • concrete failure mode
  • exact file or surface when available
  • the smallest useful fix direction or missing test

If there are no findings, say that explicitly and note any residual testing gap.

Review Workflow

  1. Identify the intended change.

    • What behavior is meant to change?
    • What behavior must remain stable?
  2. Inspect the highest-risk surfaces.

    • request and response validation
    • state transitions and persisted output
    • error handling and fallback behavior
    • env vars, IAM assumptions, dependency packaging, and template wiring
    • tests, harness coverage, and docs drift
  3. Challenge unproven assumptions.

    • Missing fields, malformed payloads, stale env vars, and partial deployments matter more than the happy path.
  4. Prefer evidence.

    • Ground findings in code, tests, runtime output, or a concrete deploy-readiness gap.
  5. Return findings in priority order.

    • bugs and release risks first
    • then fragile behavior or missing tests
    • then lower-value clarity issues only if they affect correctness

Severity Guide

  • High: likely broken behavior, data loss, or release blocker
  • Medium: works in the happy path but is fragile or untested in an important case
  • Low: clarity or maintainability issue with real defect risk
  • Question: unresolved ambiguity that could hide a defect

When not to use it

  • When performing general code style reviews
  • When the change is not a deploy candidate

Limitations

  • Findings must be grounded in code, tests, or runtime output
  • Requires priority ordering of findings

How it compares

Unlike general reviews, this process prioritizes findings over summaries and mandates grounding every observation in concrete evidence or testing gaps.

Compared to similar skills

risk-review side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
risk-review (this skill)03moNo flagsAdvanced
django-verification54moReviewIntermediate
proof-of-work16moNo flagsIntermediate
python-testing-patterns772moReviewIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry