Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Coordinate plan files by launching multiple parallel subagents for unblocked tasks. Triggers on explicit "/parallel-task" commands.
.claude/skills/jdrhyne-parallel-task/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | 80% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-21 | ✓→✗ | ▼ Worse | -31% | 0% |
| case-06 | ✓→✓ | = Same ✓ | -7% | 0% |
| case-07 | ✓→✓ | = Same ✓ | 82% | 0% |
Use Orchestration Mode. Parse plan files and delegate tasks to parallel subagents.
/parallel-task or clearly asked for parallel agent work.Extract from user request:
If sprint not provided, ask user which to run.
## Sprint 1:)### Task 1.1:)For each task, launch subagent with:
If there are multiple unblocked tasks, launch multiple subagents in parallel.
You are implementing a specific task from a development plan.
## Context
- Plan: [filename] - [sprint/phase]
- Goals: [relevant overview from plan]
- Dependencies: [prerequisites for this task]
- Related tasks: [tasks in same sprint]
- Constraints: [risks from plan]
## Your Task
**Task [ID]: [Name]**
Location: [File paths]
Description: [Full description]
Acceptance Criteria:
[List from plan]
Validation:
[Tests or verification from plan]
## Instructions
1. Examine and explore the plan, all relevant files & dependencies. Ensure ambiguities are cleared up.
2. Implement changes for all acceptance criteria
3. Keep work **atomic and committable**
4. For each file: read first, edit carefully, preserve formatting
5. Run validation if feasible
6. Return summary of:
- Files modified/created
- Changes made
- How criteria are satisfied
- Validation performed or deferred
## Important
- Be careful with paths
- Stop and describe blockers if encountered
- Focus on this specific taskAfter subagents complete:
Now, pause and read the whole plan again to see what new tasks are unblocked.
Start back at step 3. Continue launching agents on unblocked tasks in parallel.
Repeat this loop until the plan is done.
Do not stop until the plan is fully completed or you encounter a blocker.
After completing all tasks, validate the plan's success by checking the final state against the original plan. Perform available tests and ensure success. If there are any issues, address them before marking the plan as complete.
After validation, provide a summary of:
/parallel-task plan.md
/parallel-task ./plans/auth-plan.md sprint 2
/parallel-task user-profile-plan.md phase 1markdown# Sprint/Phase Run Summary ## Tasks Assigned: [N] ### Completed - Task [ID]: [Name] - [Brief summary] ### Issues - Task [ID]: [Name] - Issue: [What went wrong] - Resolution: [How resolved or what's needed] ### Blocked - Task [ID]: [Name] - Blocker: [What's preventing completion] - Next Steps: [What needs to happen] ## Overall Status [Completion summary] ## Files Modified [List of changed files] ## Next Steps [Recommendations]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 4,929 | 7,823 | +59% | 1 | 1 | 0% | 200 | 1,306 | +553% | 0 | 0 | — |
case-02 | fail→fail | 10,007 | 4,151 | -59% | 1 | 1 | 0% | 310 | 1,153 | +272% | 0 | 0 | — |
case-03 | fail→fail | 6,299 | 4,812 | -24% | 1 | 1 | 0% | 416 | 1,181 | +184% | 0 | 0 | — |
case-04 | fail→fail | 4,887 | 5,649 | +16% | 1 | 1 | 0% | 670 | 1,258 | +88% | 0 | 0 | — |
case-05 | fail→fail | 5,672 | 3,385 | -40% | 1 | 1 | 0% | 275 | 1,425 | +418% | 0 | 0 | — |
case-06 | pass→pass | 14,005 | 6,956 | -50% | 1 | 1 | 0% | 2,334 | 2,182 | -7% | 0 | 0 | — |
case-07 | pass→pass | 4,585 | 2,308 | -50% | 1 | 1 | 0% | 730 | 1,329 | +82% | 0 | 0 | — |
case-08 | pass→pass | 11,498 | 7,360 | -36% | 1 | 1 | 0% | 1,911 | 2,339 | +22% | 0 | 0 | — |
case-09 | fail→pass | 5,368 | 3,596 | -33% | 1 | 1 | 0% | 860 | 1,552 | +80% | 0 | 0 | — |
case-10 | fail→fail | 10,635 | 6,060 | -43% | 1 | 1 | 0% | 1,649 | 1,963 | +19% | 0 | 0 | — |
case-11 | pass→pass | 4,278 | 2,233 | -48% | 1 | 1 | 0% | 692 | 1,328 | +92% | 0 | 0 | — |
case-12 | fail→fail | 4,867 | 5,254 | +8% | 1 | 1 | 0% | 208 | 1,247 | +500% | 0 | 0 | — |
case-13 | fail→pass | 9,769 | 6,731 | -31% | 1 | 1 | 0% | 1,618 | 2,142 | +32% | 0 | 0 | — |
case-14 | pass→pass | 6,428 | 3,893 | -39% | 1 | 1 | 0% | 1,046 | 1,698 | +62% | 0 | 0 | — |
case-15 | fail→fail | 4,443 | 5,776 | +30% | 1 | 1 | 0% | 182 | 1,288 | +608% | 0 | 0 | — |
case-16 | pass→pass | 16,768 | 11,141 | -34% | 1 | 1 | 0% | 2,855 | 2,882 | +1% | 0 | 0 | — |
case-17 | pass→pass | 3,050 | 2,931 | -4% | 1 | 1 | 0% | 532 | 1,506 | +183% | 0 | 0 | — |
case-18 | pass→pass | 11,264 | 5,524 | -51% | 1 | 1 | 0% | 1,817 | 1,875 | +3% | 0 | 0 | — |
case-19 | pass→pass | 12,173 | 6,210 | -49% | 1 | 1 | 0% | 2,167 | 2,251 | +4% | 0 | 0 | — |
case-20 | fail→fail | 2,936 | 5,217 | +78% | 1 | 1 | 0% | 450 | 1,308 | +191% | 0 | 0 | — |
case-21 | pass→fail | 12,733 | 5,644 | -56% | 1 | 1 | 0% | 2,049 | 1,408 | -31% | 0 | 0 | — |
case-22 | pass→pass | 9,976 | 4,711 | -53% | 1 | 1 | 0% | 1,479 | 1,665 | +13% | 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 13 counted toward the lift figure. The other 9 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 +5 percentage points is the difference between those two pass rates over the 13 comparable cases. 1 case got worse with the skill loaded, and it is 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.