SkillVaultskills Browse all 1,000+ skills

Security · Version 1.3.0 · Reviewed 2026-08-02

Pull Request Security Risk Reviewer

Find and prioritize exploitable risk in security diff review and changed attack-path analysis with evidence, explicit trade-offs, and a verification plan.

4 method steps 4 documented failure modes 4 diagnostic checks 7 quality gates

Reviews a change for reachable authentication, authorization, injection, secret, deserialization, and disclosure risks. It grounds the decision in the base diff, trust boundaries, identity flows, data sensitivity, deployment context, and existing controls and explicitly prevents reporting generic weakness categories without a changed source, reachable sink, attacker capability, or control bypass.

₹199 one-time

Get this skill archive

Install in your AI coding tool

SkillVault packages this skill in the open Agent Skills format for five leading coding tools.

What this skill helps you do

  • Security diff review
  • Changed attack-path analysis
  • Security finding validation

How Pull Request Security Risk Reviewer works

You provide

Code, configuration, and the deployment trust model

It inspects

Reachable input-to-sink paths for security diff review

It decides

A changed attack-path analysis finding ranked by blast radius

You verify

Re-attempt the exploit path after remediation

What it checks first

Pull Request Security Risk Reviewer reviews a change for reachable authentication, authorization, injection, secret, deserialization, and disclosure risks. It grounds the decision in the base diff, trust boundaries, identity flows, data sensitivity, deployment context, and existing controls and explicitly prevents reporting generic weakness categories without a changed source, reachable sink, attacker capability, or control bypass. Use it when the work involves Security diff review, Changed attack-path analysis, Security finding validation.

  1. Trust boundaries and every point where untrusted input crosses one.
  2. Where authorization is enforced relative to where data is accessed.
  3. Secret handling: creation, storage, transmission, rotation, and revocation.
  4. What an attacker gains at each step, which determines whether a finding is material.

Failure modes it recognizes

  • Authorization enforced at the perimeter while internal callers reach the same data unchecked.
  • A single unparameterized query path among many parameterized ones.
  • Sensitive values written to logs or error responses.
  • A dependency vulnerability that is reachable in one code path and unreachable in the rest.

Answers it will reject

  • Reporting theoretical findings as exploitable without a demonstrated path.
  • Blocking a payload signature instead of removing the vulnerability class.
  • Treating obscurity as a control, which delays discovery without preventing exploitation.

Decision rules it applies

  • Prioritize by reachability and blast radius, not by scanner severity.
  • Fail closed on any ambiguity in an access decision.
  • Prefer eliminating the capability over sanitizing input into it.

Evidence it asks for

  • Trace input to sink and name every file and function on the path.
  • Verify the fix by attempting the original exploit path.
  • Check logs for prior exploitation before closing a finding.

The method inside

  1. Map the artifact, actors, boundaries, and invariants relevant to security diff review.
  2. Trace concrete failure or abuse paths for changed attack-path analysis; do not report checklist items without a mechanism.
  3. Prioritize security finding validation findings by impact, likelihood, confidence, and cost of correction.
  4. Recommend the smallest defensible change, then define how an independent reviewer can verify it.

Deliverables

  • Security diff review assessment
  • Changed attack-path analysis decision and action plan
  • Security finding validation verification checklist

Evidence requirements

  • Code, configuration, data flows, and trust boundaries
  • Identity, authorization, and deployment context
  • Threat model, controls, and known assumptions

Quality gates

  • Every material claim traces to supplied evidence or is labeled as a hypothesis.
  • The response follows the declared deliverable contract.
  • No execution, access, measurement, or verification is invented.
  • Secrets and personal data are redacted rather than repeated.
  • The user receives a concrete independent verification step.
  • The relevant failure modes in this domain were considered rather than only the reported symptom.
  • No listed anti-pattern was recommended as a solution.

Example task

Input

Apply the pull request security risk reviewer to our current security diff review work. We need a concrete decision, bounded changes, and evidence that the result is correct.

Expected output

Start with the base diff, trust boundaries, identity flows, data sensitivity, deployment context, and existing controls. The highest-risk failure is reporting generic weakness categories without a changed source, reachable sink, attacker capability, or control bypass. Block only concrete paths with stated preconditions, impact, confidence, and a testable remediation. Verify the result by re-attempting the exact path against the corrected revision and checking security telemetry records the denial.

Boundaries and compatibility

Ideal for

  • Security diff review: produce a decision or artifact grounded in supplied evidence.
  • Changed attack-path analysis: produce a decision or artifact grounded in supplied evidence.
  • Security finding validation: produce a decision or artifact grounded in supplied evidence.

Out of scope

  • Authorizing offensive actions against systems without permission
  • Reporting theoretical issues as exploitable without a path

Agent compatibility

  • GitHub Copilot Agent Skills
  • Cursor Agent Skills
  • Claude Code Skills
  • OpenAI Codex Skills
  • JetBrains Junie Skills

Tool policy: Advisory by default. No tools are assumed. If the host provides tools, use read-only evidence gathering unless the user explicitly approves a scoped write or execution action.