Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Launch N parallel subagents in isolated git worktrees to compete on the session task. Use when the user runs /hub:spawn or asks to start the competing agents for an initialized AgentHub session.
.claude/skills/alirezarezvani-spawn/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 173% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -43% | 0% |
Spawn N subagents that work on the same task in parallel, each in an isolated git worktree.
/hub:spawn # Spawn agents for the latest session
/hub:spawn 20260317-143022 # Spawn agents for a specific session
/hub:spawn --template optimizer # Use optimizer template for dispatch prompts
/hub:spawn --template refactorer # Use refactorer templateWhen --template <name> is provided, use the dispatch prompt from ../agenthub/references/agent-templates.md instead of the default prompt below. Available templates:
| Template | Pattern | Use Case | |----------|---------|----------| | optimizer | Edit → eval → keep/discard → repeat x10 | Performance, latency, size reduction | | refactorer | Restructure → test → iterate until green | Code quality, tech debt | | test-writer | Write tests → measure coverage → repeat | Test coverage gaps | | bug-fixer | Reproduce → diagnose → fix → verify | Bug fix with competing approaches |
When using a template, replace all {variables} with values from the session config. Assign each agent a different strategy appropriate to the template and task — diverse strategies maximize the value of parallel exploration.
.agenthub/sessions/{session-id}/config.yaml.agenthub/board/dispatch/Agent(
prompt: "You are agent-{i} in hub session {session-id}.
Your task: {task}
Read your full assignment at .agenthub/board/dispatch/{seq}-agent-{i}.md
Instructions:
1. Work in your worktree — make changes, run tests, iterate
2. Commit all changes with descriptive messages
3. Write your result summary to .agenthub/board/results/agent-{i}-result.md
Include: approach taken, files changed, metric if available, confidence level
4. Exit when done
Constraints:
- Do NOT read or modify other agents' work
- Do NOT access .agenthub/board/results/ for other agents
- Commit early and often with descriptive messages
- If you hit a dead end, commit what you have and explain in your result",
isolation: "worktree"
)running via:bashpython {skill_path}/scripts/session_manager.py --update {session-id} --state running
Tell the user:
/hub:hub-status/hub:eval| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-10 | fail→pass | 34,396 | 19,291 | -44% | 1 | 1 | 0% | 1,096 | 2,988 | +173% | 0 | 0 | — |
case-01 | fail→fail | 10,891 | 16,216 | +49% | 1 | 1 | 0% | 1,864 | 1,027 | -45% | 0 | 0 | — |
case-02 | fail→fail | 13,264 | 11,273 | -15% | 1 | 1 | 0% | 2,320 | 1,133 | -51% | 0 | 0 | — |
case-03 | fail→fail | 12,346 | 12,847 | +4% | 1 | 1 | 0% | 1,124 | 1,091 | -3% | 0 | 0 | — |
case-04 | fail→fail | 22,008 | 19,805 | -10% | 1 | 1 | 0% | 3,367 | 1,089 | -68% | 0 | 0 | — |
case-05 | fail→fail | 9,385 | 13,025 | +39% | 1 | 1 | 0% | 760 | 989 | +30% | 0 | 0 | — |
case-06 | fail→fail | 8,313 | 12,384 | +49% | 1 | 1 | 0% | 465 | 1,269 | +173% | 0 | 0 | — |
case-07 | fail→pass | 15,257 | 6,740 | -56% | 1 | 1 | 0% | 2,114 | 2,078 | -2% | 0 | 0 | — |
case-08 | fail→pass | 15,015 | 7,105 | -53% | 1 | 1 | 0% | 2,411 | 1,085 | -55% | 0 | 0 | — |
case-09 | fail→fail | 17,292 | 9,286 | -46% | 1 | 1 | 0% | 1,896 | 1,437 | -24% | 0 | 0 | — |
case-11 | fail→pass | 13,514 | 9,020 | -33% | 1 | 1 | 0% | 2,027 | 1,386 | -32% | 0 | 0 | — |
case-12 | fail→pass | 17,044 | 7,596 | -55% | 1 | 1 | 0% | 2,102 | 1,188 | -43% | 0 | 0 | — |
case-13 | fail→pass | 15,123 | 8,445 | -44% | 1 | 1 | 0% | 2,521 | 1,326 | -47% | 0 | 0 | — |
case-14 | fail→pass | 17,138 | 1,933 | -89% | 1 | 1 | 0% | 2,020 | 1,033 | -49% | 0 | 0 | — |
case-15 | fail→pass | 14,969 | 7,988 | -47% | 1 | 1 | 0% | 1,551 | 1,080 | -30% | 0 | 0 | — |
case-16 | fail→pass | 18,133 | 7,622 | -58% | 1 | 1 | 0% | 2,232 | 1,108 | -50% | 0 | 0 | — |
case-17 | pass→pass | 15,820 | 7,183 | -55% | 1 | 1 | 0% | 1,803 | 1,052 | -42% | 0 | 0 | — |
case-18 | pass→pass | 15,840 | 9,334 | -41% | 1 | 1 | 0% | 1,774 | 1,516 | -15% | 0 | 0 | — |
case-19 | pass→pass | 19,466 | 15,829 | -19% | 1 | 1 | 0% | 2,530 | 2,776 | +10% | 0 | 0 | — |
case-20 | fail→fail | 11,436 | 18,140 | +59% | 1 | 1 | 0% | 397 | 1,157 | +191% | 0 | 0 | — |
case-21 | fail→fail | 2,863 | 8,832 | +208% | 1 | 1 | 0% | 447 | 1,229 | +175% | 0 | 0 | — |
case-22 | fail→fail | 5,890 | 15,501 | +163% | 1 | 1 | 0% | 1,200 | 915 | -24% | 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, and 12 counted toward the lift figure. The other 10 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 +41 percentage points is the difference between those two pass rates over the 12 comparable cases. 3 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/11/2026 | +27% |
Other measured skills in the registry, with their headline benchmark lift.