Differential Security Review

What it does

When your agent uses it

• Reviewing a PR or commit for security vulnerabilities
• Checking whether a change reintroduces a bug that was fixed before
• Working out what else a change could break
• Finding modified code that no test covers

Example prompts

• “Do a security review of this diff against main.”
• “Check whether this commit reverts the earlier fix for the auth bypass, using git blame.”
• “Which functions touched by this PR have no tests, and how many callers do they have?”

Requirements

• A Git repository containing the change to review
• Pre-approved tools (allowed-tools): Read, Write, Grep, Glob, Bash

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

1. 1Risk-First: Focus on auth, crypto, value transfer, external calls
2. 2Evidence-Based: Every finding backed by git history, line numbers, attack scenarios
3. 3Adaptive: Scale to codebase size (SMALL/MEDIUM/LARGE)
4. 4Honest: Explicitly state coverage limits and confidence level
5. 5Output-Driven: Always generate comprehensive markdown report file

What it can do on your machine

Read from SKILL.md and the folder at commit 82fe822. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Write
  • Grep
  • Glob
  • Bash

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md.

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Differential Security Review loads about 1.8k tokens when it runs. Until then it costs about 124 tokens; SKILL.md has 583 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~124

When it runs · the whole SKILL.md, loaded when a task matches

~1.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Read, Write, Grep, Glob, Bash

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from trailofbits/skills at commit 82fe822, republished under its CC-BY-SA-4.0 licence (© trailofbits). 583 words, ~1,835 tokens.

name

differential-review

description

Performs security-focused differential review of code changes. Adapts analysis depth to codebase size, uses git blame for context, calculates blast radius by counting callers, checks test coverage of modified code, and generates a markdown report. Use when reviewing a PR, commit, or diff for security vulnerabilities, checking whether a change re-introduces a previously fixed bug, asking what else a change could break, or finding which modified code has no test covering it.

allowed-tools

Read, Write, Grep, Glob, Bash

Pre-Analysis → Phase 0: Triage → Phase 1: Code Analysis → Phase 2: Test Coverage
    ↓              ↓                    ↓                        ↓
Phase 3: Blast Radius → Phase 4: Deep Context → Phase 5: Adversarial → Phase 6: Report

├─ Need detailed phase-by-phase methodology?
│  └─ Read: methodology.md
│     (Pre-Analysis + Phases 0-4: triage, code analysis, test coverage, blast radius)
│
├─ Analyzing HIGH RISK change?
│  ├─ Read: adversarial.md
│  │  (Phase 5: Attacker modeling, exploit scenarios, exploitability rating)
│  └─ Or delegate to: differential-review:adversarial-modeler agent
│     (Autonomous attacker modeling with concrete exploit scenarios)
│
├─ Writing the final report?
│  └─ Read: reporting.md
│     (Phase 6: Report structure, templates, formatting guidelines)
│
├─ Looking for specific vulnerability patterns?
│  └─ Read: patterns.md
│     (Regressions, reentrancy, access control, overflow, etc.)
│
└─ Quick triage only?
   └─ Use Quick Reference above, skip detailed docs

Input: 5 file PR, 2 HIGH RISK files
Strategy: Use Quick Reference
1. Classify risk level per file (2 HIGH, 3 LOW)
2. Focus on 2 HIGH files only
3. Git blame removed code
4. Generate minimal report
Time: ~30 minutes

Input: 80 files, 12 HIGH RISK changes
Strategy: FOCUSED (see methodology.md)
1. Full workflow on HIGH RISK files
2. Surface scan on MEDIUM
3. Skip LOW risk files
4. Complete report with all sections
Time: ~3-4 hours

Input: 450 files, auth system rewrite
Strategy: SURGICAL + audit-context-building
1. Baseline context with audit-context-building
2. Deep analysis on auth changes only
3. Blast radius analysis
4. Adversarial modeling
5. Comprehensive report
Time: ~6-8 hours

© trailofbits, CC-BY-SA-4.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 6 other files (assets) in plugins/differential-review/skills/differential-review of trailofbits/skills.

• SKILL.md
• adversarial.md
• agents/openai.yaml
• assets/trail-of-bits-mark.svg
• methodology.md
• patterns.md
• reporting.md

Open the folder on GitHubat commit 82fe822