Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Parallel Agent Type Contracts
.claude/skills/parallel-agent-contracts/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 7 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -48% | 0% |
When launching parallel agents for code implementation, prevent type duplication.
markdown## Before Marking Complete Run verification: \`\`\`bash npx tsc --noEmit 2>&1 | head -20 \`\`\` If ANY type errors exist, fix them before completing.
markdown## Before Creating Any Type/Interface First check if it exists: \`\`\`bash grep -r "interface YourTypeName\|type YourTypeName" src/ \`\`\` If found, import it. NEVER duplicate existing types.
Include relevant entries from this map in agent prompts:
| Type | Owner File | Import From | |------|-----------|-------------| | NormalizedTool | src/sdk/agent.ts | './agent' | | ToolCall | src/sdk/agent.ts | './agent' | | ToolResult | src/sdk/agent.ts | './agent' | | ToolDefinition | src/sdk/agent.ts | './agent' | | Message | src/sdk/types.ts | './types' | | ContentBlock | src/sdk/types.ts | './types' | | TokenUsage | src/sdk/types.ts | './types' | | ProviderAdapter | src/sdk/providers/index.ts | './providers' | | RiggClient | src/sdk/client.ts | './client' |
When spawning implementation agents:
markdown# Task: [Description] ## Type Ownership (DO NOT recreate) - [List relevant types from canonical map] ## Before Creating New Types Run: `grep -r "interface TypeName" src/` - if exists, import it. ## Before Marking Complete Run: `npx tsc --noEmit 2>&1 | head -20` Fix all type errors before completing. ## Your Implementation [Actual task description]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 15,203 | 16,176 | +6% | 1 | 1 | 0% | 2,695 | 2,957 | +10% | 0 | 0 | — |
case-01 | fail→pass | 13,129 | 5,357 | -59% | 1 | 1 | 0% | 2,180 | 1,615 | -26% | 0 | 0 | — |
case-02 | fail→pass | 14,668 | 7,061 | -52% | 1 | 1 | 0% | 2,391 | 2,150 | -10% | 0 | 0 | — |
case-03 | fail→pass | 12,190 | 7,026 | -42% | 1 | 1 | 0% | 2,241 | 1,953 | -13% | 0 | 0 | — |
case-04 | pass→pass | 14,042 | 11,761 | -16% | 1 | 1 | 0% | 2,816 | 3,112 | +11% | 0 | 0 | — |
case-06 | pass→pass | 7,546 | 6,600 | -13% | 1 | 1 | 0% | 1,612 | 2,031 | +26% | 0 | 0 | — |
case-07 | fail→fail | 11,706 | 11,144 | -5% | 1 | 1 | 0% | 2,167 | 2,519 | +16% | 0 | 0 | — |
case-08 | pass→pass | 9,199 | 17,519 | +90% | 1 | 1 | 0% | 1,682 | 3,260 | +94% | 0 | 0 | — |
case-09 | pass→pass | 10,462 | 9,575 | -8% | 1 | 1 | 0% | 1,834 | 2,198 | +20% | 0 | 0 | — |
case-10 | fail→pass | 8,895 | 3,702 | -58% | 1 | 1 | 0% | 1,593 | 1,286 | -19% | 0 | 0 | — |
case-11 | fail→pass | 12,699 | 3,086 | -76% | 1 | 1 | 0% | 2,280 | 1,176 | -48% | 0 | 0 | — |
case-12 | fail→pass | 12,258 | 5,054 | -59% | 1 | 1 | 0% | 2,325 | 1,488 | -36% | 0 | 0 | — |
case-13 | fail→fail | 17,486 | 2,028 | -88% | 1 | 1 | 0% | 1,218 | 869 | -29% | 0 | 0 | — |
case-14 | fail→pass | 9,703 | 2,213 | -77% | 1 | 1 | 0% | 1,671 | 930 | -44% | 0 | 0 | — |
case-15 | fail→fail | 7,474 | 2,061 | -72% | 1 | 1 | 0% | 1,493 | 847 | -43% | 0 | 0 | — |
case-16 | fail→pass | 27,689 | 11,743 | -58% | 1 | 1 | 0% | 998 | 2,651 | +166% | 0 | 0 | — |
case-17 | pass→pass | 13,605 | 10,702 | -21% | 1 | 1 | 0% | 2,169 | 2,217 | +2% | 0 | 0 | — |
case-18 | pass→pass | 4,652 | 3,539 | -24% | 1 | 1 | 0% | 738 | 890 | +21% | 0 | 0 | — |
case-19 | fail→pass | 7,775 | 4,549 | -41% | 1 | 1 | 0% | 1,278 | 1,280 | +0% | 0 | 0 | — |
case-20 | pass→pass | 10,086 | 8,745 | -13% | 1 | 1 | 0% | 1,748 | 2,023 | +16% | 0 | 0 | — |
case-21 | fail→pass | 15,576 | 12,474 | -20% | 1 | 1 | 0% | 2,646 | 2,629 | -1% | 0 | 0 | — |
case-22 | pass→pass | 15,913 | 3,111 | -80% | 1 | 1 | 0% | 1,662 | 1,058 | -36% | 0 | 0 | — |
case-23 | fail→fail | 3,307 | 2,097 | -37% | 1 | 1 | 0% | 586 | 968 | +65% | 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 21 counted toward the lift figure. The other 2 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 +43 percentage points is the difference between those two pass rates over the 21 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 7/29/2026 | +57% |
Other measured skills in the registry, with their headline benchmark lift.