Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Review contracts against your organization's negotiation playbook, flagging deviations and generating redline suggestions. Use when reviewing vendor contracts, customer agreements, or any commercial agreement where you need clause-by-clause analysis against standard positions.
.claude/skills/vythanhtra-legal-contract-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 57% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 5% | 0% |
You are a contract review assistant for an in-house legal team. You analyze contracts against the organization's negotiation playbook, identify deviations, classify their severity, and generate actionable redline suggestions.
Before reviewing any contract, check for a configured playbook in the user's local settings.
If no playbook is available:
Aligns with or is better than standard position. No negotiation needed.
Falls outside standard position but within negotiable range. Requires attention and likely negotiation.
Falls outside acceptable range or poses material risk. Requires senior counsel review or business decision-maker sign-off.
**Clause**: [Section reference and clause name]
**Current language**: "[exact quote from the contract]"
**Proposed redline**: "[specific alternative language]"
**Rationale**: [1-2 sentences explaining why]
**Priority**: [Must-have / Should-have / Nice-to-have]
**Fallback**: [Alternative position if primary redline is rejected]Strategy: Lead with Tier 1. Trade Tier 3 concessions to secure Tier 2 wins. Never concede on Tier 1 without escalation.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | fail→pass | 14,262 | 10,956 | -23% | 1 | 1 | 0% | 2,385 | 2,771 | +16% | 0 | 0 | — |
case-01 | fail→fail | 34,611 | 8,965 | -74% | 1 | 1 | 0% | 6,232 | 2,418 | -61% | 0 | 0 | — |
case-02 | fail→fail | 13,288 | 29,158 | +119% | 1 | 1 | 0% | 1,339 | 5,315 | +297% | 0 | 0 | — |
case-03 | fail→pass | 10,315 | 8,007 | -22% | 1 | 1 | 0% | 1,419 | 2,233 | +57% | 0 | 0 | — |
case-04 | fail→pass | 15,591 | 10,381 | -33% | 1 | 1 | 0% | 2,612 | 2,642 | +1% | 0 | 0 | — |
case-05 | pass→pass | 7,353 | 5,331 | -27% | 1 | 1 | 0% | 1,254 | 1,740 | +39% | 0 | 0 | — |
case-06 | fail→fail | 14,347 | 10,780 | -25% | 1 | 1 | 0% | 2,603 | 2,788 | +7% | 0 | 0 | — |
case-07 | pass→pass | 9,621 | 13,975 | +45% | 1 | 1 | 0% | 1,553 | 2,367 | +52% | 0 | 0 | — |
case-08 | fail→pass | 9,749 | 7,176 | -26% | 1 | 1 | 0% | 1,819 | 2,242 | +23% | 0 | 0 | — |
case-09 | fail→pass | 11,699 | 6,976 | -40% | 1 | 1 | 0% | 1,905 | 2,007 | +5% | 0 | 0 | — |
case-10 | pass→pass | 10,124 | 7,978 | -21% | 1 | 1 | 0% | 1,788 | 2,331 | +30% | 0 | 0 | — |
case-12 | pass→pass | 11,168 | 8,645 | -23% | 1 | 1 | 0% | 1,941 | 2,440 | +26% | 0 | 0 | — |
case-13 | pass→pass | 10,298 | 8,907 | -14% | 1 | 1 | 0% | 1,693 | 2,491 | +47% | 0 | 0 | — |
case-14 | pass→pass | 12,243 | 9,764 | -20% | 1 | 1 | 0% | 2,008 | 2,519 | +25% | 0 | 0 | — |
case-15 | pass→pass | 19,545 | 8,660 | -56% | 1 | 1 | 0% | 1,512 | 2,611 | +73% | 0 | 0 | — |
case-16 | pass→pass | 13,949 | 11,190 | -20% | 1 | 1 | 0% | 2,273 | 2,721 | +20% | 0 | 0 | — |
case-17 | pass→pass | 10,588 | 10,449 | -1% | 1 | 1 | 0% | 1,933 | 2,714 | +40% | 0 | 0 | — |
case-18 | pass→pass | 8,089 | 7,705 | -5% | 1 | 1 | 0% | 1,515 | 2,345 | +55% | 0 | 0 | — |
case-19 | pass→pass | 8,521 | 3,678 | -57% | 1 | 1 | 0% | 1,309 | 1,578 | +21% | 0 | 0 | — |
case-20 | fail→pass | 16,757 | 12,538 | -25% | 1 | 1 | 0% | 3,081 | 3,112 | +1% | 0 | 0 | — |
case-21 | fail→fail | 7,421 | 10,175 | +37% | 1 | 1 | 0% | 1,071 | 2,419 | +126% | 0 | 0 | — |
case-22 | fail→fail | 17,518 | 10,858 | -38% | 1 | 1 | 0% | 3,686 | 3,090 | -16% | 0 | 0 | — |
case-23 | pass→pass | 10,154 | 4,859 | -52% | 1 | 1 | 0% | 1,601 | 1,678 | +5% | 0 | 0 | — |
case-24 | pass→pass | 8,908 | 7,888 | -11% | 1 | 1 | 0% | 1,591 | 2,362 | +48% | 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. 24 cases were attempted. The headline lift of +25 percentage points is the difference between those two pass rates over the 24 comparable cases.
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.