Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Guides through Trail of Bits' 5-step secure development workflow. Runs Slither scans, checks special features (upgradeability/ERC conformance/token integration), generates visual security diagrams, helps document security properties for fuzzing/verification, and reviews manual security areas.
.claude/skills/trailofbits-secure-workflow-guide/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 87% | 0% |
Guides through Trail of Bits' secure development workflow - a 5-step process to enhance smart contract security throughout development.
Use this: On every check-in, before deployment, or when you want a security review
Covers a security workflow including:
Run Slither with 70+ built-in detectors to find common vulnerabilities:
Goal: Clean Slither report or documented triages
Detect and validate applicable features:
Note: Only runs checks that apply to your codebase
Generate 3 security diagrams:
Review each diagram for security concerns
Help document critical security properties:
Then set up testing:
Note: Most important activity for security
Analyze areas automated tools miss:
Search codebase for these patterns and flag risks
For detailed instructions, commands, and explanations for each step, see WORKFLOW_STEPS.md.
When invoked, I will:
Adapts based on:
| Rationalization | Why It's Wrong | Required Action | |-----------------|----------------|-----------------| | "Slither not available, I'll check manually" | Manual checking misses 70+ detector patterns | Install and run Slither, or document why it's blocked | | "Can't generate diagrams, I'll describe the architecture" | Descriptions aren't visual - diagrams reveal patterns text misses | Execute slither --print commands, generate actual visual outputs | | "No upgrades detected, skip upgradeability checks" | Proxies and upgrades are often implicit or planned | Verify with codebase search before skipping Step 2 checks | | "Not a token, skip ERC checks" | Tokens can be integrated without obvious ERC inheritance | Check for token interactions, transfers, balances before skipping | | "Can't set up Echidna now, suggesting it for later" | Property-based testing is Step 4, not optional | Document properties now, set up fuzzing infrastructure | | "No DeFi interactions, skip oracle/flash loan checks" | DeFi patterns appear in unexpected places (price feeds, external calls) | Complete Step 5 manual review, search codebase for patterns | | "This step doesn't apply to my project" | "Not applicable" without verification = missed vulnerabilities | Verify with explicit codebase search before declaring N/A | | "I'll provide generic security advice instead of running workflow" | Generic advice isn't actionable, workflow finds specific issues | Execute all 5 steps, generate project-specific findings with file:line references |
When I complete the workflow, you'll get a comprehensive security report covering:
For a complete example workflow report, see EXAMPLE_REPORT.md.
Security Report:
Action Plan:
Workflow Checklist:
Trail of Bits Resources:
Other Security:
Let me know when you're ready and I'll run through the workflow with your codebase!
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | pass→pass | 9,844 | 6,338 | -36% | 1 | 1 | 0% | 1,811 | 2,466 | +36% | 0 | 0 | — |
case-10 | pass→pass | 9,778 | 11,958 | +22% | 1 | 1 | 0% | 1,761 | 2,540 | +44% | 0 | 0 | — |
case-11 | pass→pass | 9,715 | 6,235 | -36% | 1 | 1 | 0% | 1,589 | 2,527 | +59% | 0 | 0 | — |
case-12 | fail→pass | 14,789 | 6,874 | -54% | 1 | 1 | 0% | 2,195 | 2,551 | +16% | 0 | 0 | — |
case-13 | pass→fail | 6,210 | 4,134 | -33% | 1 | 1 | 0% | 1,191 | 2,091 | +76% | 0 | 0 | — |
case-01 | fail→fail | 23,473 | 10,953 | -53% | 1 | 1 | 0% | 2,925 | 2,626 | -10% | 0 | 0 | — |
case-02 | fail→fail | 9,942 | 70,179 | +606% | 1 | 1 | 0% | 1,152 | 2,512 | +118% | 0 | 0 | — |
case-03 | fail→fail | 16,201 | 9,933 | -39% | 1 | 1 | 0% | 1,022 | 2,399 | +135% | 0 | 0 | — |
case-04 | pass→pass | 13,016 | 11,206 | -14% | 1 | 1 | 0% | 2,803 | 3,831 | +37% | 0 | 0 | — |
case-05 | pass→pass | 20,415 | 18,257 | -11% | 1 | 1 | 0% | 4,778 | 5,815 | +22% | 0 | 0 | — |
case-06 | pass→pass | 8,265 | 15,369 | +86% | 1 | 1 | 0% | 1,600 | 4,220 | +164% | 0 | 0 | — |
case-07 | pass→pass | 12,433 | 7,598 | -39% | 1 | 1 | 0% | 2,111 | 2,663 | +26% | 0 | 0 | — |
case-08 | fail→pass | 9,804 | 4,910 | -50% | 1 | 1 | 0% | 1,572 | 2,228 | +42% | 0 | 0 | — |
case-14 | pass→pass | 9,334 | 3,120 | -67% | 1 | 1 | 0% | 1,740 | 1,926 | +11% | 0 | 0 | — |
case-15 | pass→pass | 5,058 | 2,799 | -45% | 1 | 1 | 0% | 880 | 1,800 | +105% | 0 | 0 | — |
case-16 | pass→pass | 5,623 | 3,468 | -38% | 1 | 1 | 0% | 1,075 | 1,712 | +59% | 0 | 0 | — |
case-17 | fail→pass | 6,784 | 1,761 | -74% | 1 | 1 | 0% | 1,245 | 1,647 | +32% | 0 | 0 | — |
case-18 | pass→pass | 6,169 | 2,291 | -63% | 1 | 1 | 0% | 1,185 | 1,683 | +42% | 0 | 0 | — |
case-19 | fail→pass | 9,715 | 2,678 | -72% | 1 | 1 | 0% | 1,698 | 1,886 | +11% | 0 | 0 | — |
case-20 | fail→pass | 6,332 | 3,792 | -40% | 1 | 1 | 0% | 1,083 | 2,025 | +87% | 0 | 0 | — |
case-21 | fail→pass | 8,751 | 5,282 | -40% | 1 | 1 | 0% | 1,574 | 2,435 | +55% | 0 | 0 | — |
case-22 | fail→pass | 20,126 | 3,914 | -81% | 1 | 1 | 0% | 1,225 | 2,193 | +79% | 0 | 0 | — |
case-23 | pass→pass | 9,985 | 3,021 | -70% | 1 | 1 | 0% | 1,759 | 1,936 | +10% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted, and 22 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +26 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.