Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Review business contracts for risk identification including liability clauses, IP ownership, termination terms, and payment conditions. Use this skill when the user needs a practical contract risk assessment for vendor agreements, partnership contracts, or service agreements — even if they say 'review this contract', 'what should I watch out for', 'is this agreement fair', or 'negotiate better terms'.
.claude/skills/asgard-ai-platform-ops-contract-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-09 | ✓→✗ | ▼ Worse | 25% | 0% |
IRON LAW: Read the Entire Contract, Not Just the Commercial Terms
Price and scope are negotiated carefully. Liability, indemnification,
termination, and IP ownership are often accepted on autopilot.
These "back-of-contract" clauses are where the real risk lives.
A great price with unlimited liability is a terrible deal.1. Parties & Authority
2. Scope & Deliverables
3. Payment Terms
4. Term & Termination
5. Liability & Indemnification
6. Intellectual Property
7. Confidentiality
8. Dispute Resolution
| Risk Level | Characteristics | Action | |-----------|----------------|--------| | 🟢 Low | Market-standard terms, balanced obligations | Sign | | 🟡 Medium | Some one-sided clauses, manageable risk | Negotiate specific clauses | | 🔴 High | Unlimited liability, no exit, one-sided IP | Don't sign without major revisions |
markdown# Contract Review: {Agreement Type} with {Counterparty} ## Overview - Type: {service/vendor/partnership/license} - Value: {$X} - Term: {duration} ## Risk Assessment | Area | Risk | Issue | Recommendation | |------|------|-------|---------------| | Scope | 🟢/🟡/🔴 | {finding} | {action} | | Payment | 🟢/🟡/🔴 | {finding} | {action} | | Termination | 🟢/🟡/🔴 | {finding} | {action} | | Liability | 🟢/🟡/🔴 | {finding} | {action} | | IP | 🟢/🟡/🔴 | {finding} | {action} | | Confidentiality | 🟢/🟡/🔴 | {finding} | {action} | ## Red Flags 1. {specific clause + why it's risky} ## Negotiation Priorities 1. {most important change to request} 2. {second priority} 3. {third priority} ## Overall Risk: 🟢/🟡/🔴 {Summary recommendation: sign / negotiate / reject}
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 17,129 | 25,404 | +48% | 1 | 1 | 0% | 2,980 | 4,116 | +38% | 0 | 0 | — |
case-02 | fail→pass | 23,075 | 24,527 | +6% | 1 | 1 | 0% | 3,614 | 5,146 | +42% | 0 | 0 | — |
case-03 | fail→pass | 25,440 | 15,305 | -40% | 1 | 1 | 0% | 4,167 | 3,476 | -17% | 0 | 0 | — |
case-04 | fail→fail | 22,181 | 9,644 | -57% | 1 | 1 | 0% | 3,343 | 2,680 | -20% | 0 | 0 | — |
case-05 | fail→fail | 52,682 | 45,206 | -14% | 1 | 1 | 0% | 8,221 | 8,116 | -1% | 0 | 0 | — |
case-06 | pass→pass | 15,871 | 20,458 | +29% | 1 | 1 | 0% | 2,284 | 3,629 | +59% | 0 | 0 | — |
case-07 | pass→pass | 16,973 | 11,115 | -35% | 1 | 1 | 0% | 2,944 | 3,030 | +3% | 0 | 0 | — |
case-08 | pass→pass | 17,456 | 12,867 | -26% | 1 | 1 | 0% | 2,556 | 2,992 | +17% | 0 | 0 | — |
case-09 | pass→fail | 14,885 | 12,174 | -18% | 1 | 1 | 0% | 2,514 | 3,145 | +25% | 0 | 0 | — |
case-10 | fail→fail | 16,018 | 12,566 | -22% | 1 | 1 | 0% | 2,733 | 3,167 | +16% | 0 | 0 | — |
case-11 | pass→pass | 15,742 | 10,028 | -36% | 1 | 1 | 0% | 2,290 | 2,764 | +21% | 0 | 0 | — |
case-12 | pass→pass | 18,791 | 14,856 | -21% | 1 | 1 | 0% | 2,709 | 3,538 | +31% | 0 | 0 | — |
case-13 | pass→fail | 14,766 | 12,522 | -15% | 1 | 1 | 0% | 2,515 | 3,275 | +30% | 0 | 0 | — |
case-14 | pass→pass | 15,419 | 11,792 | -24% | 1 | 1 | 0% | 2,419 | 3,051 | +26% | 0 | 0 | — |
case-15 | pass→pass | 15,589 | 13,613 | -13% | 1 | 1 | 0% | 2,552 | 3,251 | +27% | 0 | 0 | — |
case-16 | pass→pass | 24,050 | 17,702 | -26% | 1 | 1 | 0% | 3,398 | 3,500 | +3% | 0 | 0 | — |
case-22 | fail→pass | 15,223 | 13,640 | -10% | 1 | 1 | 0% | 2,566 | 3,354 | +31% | 0 | 0 | — |
case-17 | pass→pass | 16,838 | 18,092 | +7% | 1 | 1 | 0% | 2,621 | 3,679 | +40% | 0 | 0 | — |
case-18 | pass→pass | 15,366 | 12,998 | -15% | 1 | 1 | 0% | 2,400 | 3,201 | +33% | 0 | 0 | — |
case-19 | pass→pass | 16,244 | 11,045 | -32% | 1 | 1 | 0% | 2,809 | 3,002 | +7% | 0 | 0 | — |
case-20 | pass→pass | 17,588 | 13,065 | -26% | 1 | 1 | 0% | 2,650 | 3,277 | +24% | 0 | 0 | — |
case-21 | fail→fail | 19,538 | 15,878 | -19% | 1 | 1 | 0% | 3,215 | 3,354 | +4% | 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. 22 cases were attempted. The headline lift of +9 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.