Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Context management specialist for multi-agent workflows and long-running tasks. Use when coordinating multiple agents, preserving context across sessions, or managing complex project state. Focuses on context compression and distribution.
.claude/skills/aiskillstore-context-management/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -64% | 0% |
This skill manages context across multi-agent workflows and long-running tasks, ensuring coherent state preservation and efficient context distribution.
Capture the key context from this conversationPrepare context for the next agent in this workflowCreate a context checkpoint for this project milestonereferences/patterns.md - Context management patterns, handoff templates, and multi-agent coordination| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 11,414 | 7,090 | -38% | 1 | 1 | 0% | 1,673 | 1,552 | -7% | 0 | 0 | — |
case-02 | fail→pass | 12,683 | 8,821 | -30% | 1 | 1 | 0% | 2,069 | 1,643 | -21% | 0 | 0 | — |
case-03 | pass→pass | 10,833 | 4,745 | -56% | 1 | 1 | 0% | 1,635 | 1,240 | -24% | 0 | 0 | — |
case-04 | pass→pass | 14,633 | 8,674 | -41% | 1 | 1 | 0% | 2,268 | 1,874 | -17% | 0 | 0 | — |
case-05 | pass→pass | 13,207 | 11,779 | -11% | 1 | 1 | 0% | 2,148 | 2,234 | +4% | 0 | 0 | — |
case-06 | pass→pass | 12,976 | 9,844 | -24% | 1 | 1 | 0% | 2,098 | 1,999 | -5% | 0 | 0 | — |
case-07 | pass→pass | 12,396 | 10,460 | -16% | 1 | 1 | 0% | 1,936 | 1,974 | +2% | 0 | 0 | — |
case-08 | pass→pass | 5,344 | 5,267 | -1% | 1 | 1 | 0% | 807 | 1,207 | +50% | 0 | 0 | — |
case-09 | pass→pass | 7,430 | 5,826 | -22% | 1 | 1 | 0% | 1,144 | 1,396 | +22% | 0 | 0 | — |
case-10 | pass→pass | 10,895 | 7,048 | -35% | 1 | 1 | 0% | 1,654 | 1,432 | -13% | 0 | 0 | — |
case-11 | fail→pass | 14,092 | 6,100 | -57% | 1 | 1 | 0% | 2,339 | 1,447 | -38% | 0 | 0 | — |
case-12 | pass→pass | 19,322 | 6,692 | -65% | 1 | 1 | 0% | 2,016 | 1,408 | -30% | 0 | 0 | — |
case-13 | pass→pass | 10,997 | 8,532 | -22% | 1 | 1 | 0% | 1,627 | 1,774 | +9% | 0 | 0 | — |
case-14 | pass→pass | 14,003 | 11,803 | -16% | 1 | 1 | 0% | 1,927 | 2,046 | +6% | 0 | 0 | — |
case-15 | fail→pass | 19,765 | 1,868 | -91% | 1 | 1 | 0% | 1,424 | 662 | -54% | 0 | 0 | — |
case-16 | fail→pass | 25,445 | 1,979 | -92% | 1 | 1 | 0% | 2,005 | 728 | -64% | 0 | 0 | — |
case-17 | pass→pass | 8,406 | 3,034 | -64% | 1 | 1 | 0% | 1,267 | 945 | -25% | 0 | 0 | — |
case-18 | fail→pass | 16,414 | 12,620 | -23% | 1 | 1 | 0% | 2,080 | 1,586 | -24% | 0 | 0 | — |
case-19 | pass→pass | 9,020 | 6,237 | -31% | 1 | 1 | 0% | 1,246 | 1,386 | +11% | 0 | 0 | — |
case-20 | pass→pass | 15,493 | 12,297 | -21% | 1 | 1 | 0% | 2,614 | 2,555 | -2% | 0 | 0 | — |
case-21 | pass→pass | 46,116 | 17,736 | -62% | 1 | 1 | 0% | 3,514 | 3,135 | -11% | 0 | 0 | — |
case-22 | pass→pass | 9,604 | 13,044 | +36% | 1 | 1 | 0% | 1,597 | 1,917 | +20% | 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 +27 percentage points is the difference between those two pass rates over the 22 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.