How this skill is triggered — by the user, by Claude, or both
Slash command
/power-trio:artifact-schemasThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Reference documentation for Power Trio artifact formats.
Reference documentation for Power Trio artifact formats.
Each phase produces a structured artifact that serves as input to subsequent phases. These artifacts are written to the .power-trio/ directory.
.power-trio/
├── plan.report.md # Phase 1 output
├── implementation.report.md # Phase 2 output
└── release.report.md # Phase 3 output
Produced by: Phase 1 (Plan) Consumed by: Phase 2 (Build)
# Plan Report
## Task
[Original user request - verbatim or summarized]
## Goals
- Primary: [Main objective to achieve]
- Secondary: [Additional objectives, if any]
## Constraints
- [Technical constraint]
- [Scope constraint]
- [Dependency constraint]
## Implementation Steps
1. [Step with clear deliverable]
2. [Step with clear deliverable]
3. [Step with clear deliverable]
## Success Criteria
- [ ] [Testable criterion 1]
- [ ] [Testable criterion 2]
- [ ] [Testable criterion 3]
## Risks
- [Risk]: [Mitigation strategy]
- [Risk]: [Mitigation strategy]
## Committee Approval
- Planner: APPROVE
- Architect: APPROVE
- Critic: APPROVE
## Iterations
[N] iterations to reach consensus
## Open Questions
None
---
Generated by Power Trio Phase 1: Plan
Timestamp: [ISO 8601]
| Field | Required | Description |
|---|---|---|
| Task | Yes | Original request |
| Goals | Yes | At least primary goal |
| Constraints | No | May be empty |
| Implementation Steps | Yes | At least one step |
| Success Criteria | Yes | At least one criterion |
| Risks | No | May be empty |
| Committee Approval | Yes | All three votes |
| Open Questions | Yes | Must be "None" for approval |
Produced by: Phase 2 (Build Consumed by: Phase 3 (Review)
# Implementation Report
## Task
[From plan.report]
## Files Changed
| File | Change Type | Description |
| ----------------- | ----------- | -------------------- |
| path/to/file.ext | Added | New file for X |
| path/to/other.ext | Modified | Updated Y |
| path/to/old.ext | Deleted | Removed deprecated Z |
## Lines of Code
- Added: [N]
- Modified: [N]
- Deleted: [N]
- Total: [N]
## Test Results
- Test files: [list of test file paths]
- Test cases: [N] total
- Passing: [N]
- Coverage: [N]% (if available)
- Command: `[exact command to run tests]`
## Deviations from Plan
- None
OR
- [Deviation]: [Justification for why it was necessary]
## Dependencies
- New dependencies: [list] or None
- External APIs: [list] or None
## Sensitive Areas Touched
- Authentication: Yes/No
- Payments: Yes/No
- Cryptography: Yes/No
- User data: Yes/No
- Public API: Yes/No
## Committee Approval
- Test Author: Tests approved
- Engineer: Implementation complete
- Senior Engineer: APPROVE
## Iterations
- Test iterations: [N]
- Implementation iterations: [N]
---
Generated by Power Trio Phase 2: Build
Timestamp: [ISO 8601]
| Field | Required | Description |
|---|---|---|
| Task | Yes | From plan.report |
| Files Changed | Yes | At least one file |
| Lines of Code | Yes | All four counts |
| Test Results | Yes | All fields |
| Deviations from Plan | Yes | "None" or list |
| Sensitive Areas Touched | Yes | All five booleans |
| Committee Approval | Yes | All three statuses |
These fields are used to determine if Phase 3 should run:
Produced by: Phase 3 (Review) Consumed by: Final output / user
# Release Report
## Implementation Reviewed
.power-trio/implementation.report.md
## Security Review
### Findings
| Severity | Issue | Classification | Rationale |
| -------- | ------ | -------------- | --------- |
| CRITICAL | [desc] | BLOCKING | [why] |
| HIGH | [desc] | BLOCKING | [why] |
| MEDIUM | [desc] | DEFERRED | [why] |
| LOW | [desc] | DEFERRED | [why] |
### Summary
- Critical: [N] (Blocking: [N], Deferred: [N])
- High: [N] (Blocking: [N], Deferred: [N])
- Medium: [N] (Blocking: [N], Deferred: [N])
- Low: [N] (Blocking: [N], Deferred: [N])
## Performance Review
### Findings
| Severity | Issue | Classification | Rationale |
| -------- | ------ | -------------- | --------- |
| HIGH | [desc] | DEFERRED | [why] |
| MEDIUM | [desc] | DEFERRED | [why] |
### Summary
- Critical: [N] (Blocking: [N], Deferred: [N])
- High: [N] (Blocking: [N], Deferred: [N])
- Medium: [N] (Blocking: [N], Deferred: [N])
- Low: [N] (Blocking: [N], Deferred: [N])
## Blocking Issues
```json
[]
```
OR
[
{
"source": "security",
"severity": "CRITICAL",
"issue": "Description of the issue",
"fix_required": "Specific fix that must be applied"
}
]
[
{
"source": "performance",
"severity": "MEDIUM",
"issue": "Description of the issue",
"follow_up": "Timeline or trigger for addressing"
}
]
APPROVED_TO_SHIP
OR
BLOCKED
[Product Manager's explanation of the decision]
Generated by Power Trio Phase 3: Review Timestamp: [ISO 8601] Security findings: [N] Performance findings: [N] Blocking: [N] Deferred: [N]
### Required Fields
| Field | Required | Description |
|-------|----------|-------------|
| Implementation Reviewed | Yes | Path to implementation.report |
| Security Review | Yes | Findings table and summary |
| Performance Review | Yes | Findings table and summary |
| Blocking Issues | Yes | JSON array (may be empty) |
| Deferred Issues | Yes | JSON array (may be empty) |
| Verdict | Yes | APPROVED_TO_SHIP or BLOCKED |
| Rationale | Yes | Explanation |
### JSON Schemas
**Blocking Issue:**
```json
{
"source": "security" | "performance",
"severity": "CRITICAL" | "HIGH" | "MEDIUM" | "LOW",
"issue": "string",
"fix_required": "string"
}
Deferred Issue:
{
"source": "security" | "performance",
"severity": "CRITICAL" | "HIGH" | "MEDIUM" | "LOW",
"issue": "string",
"follow_up": "string"
}
npx claudepluginhub thesimonho/artificial-jellybeans --plugin power-trioReviews planning artifacts like requirements, data models, API contracts for gaps in analysis and design phases. Classifies issues by severity for quality gates before proceeding.
Executes implementation plans phase-by-phase: dispatches subagents per task, reviews once per phase with code-review skill, loads phases just-in-time, prints full outputs for transparency.
Decomposes epics into trackable, right-sized tasks with Size, Urgency, Risk, ROI, Blast Radius, LOE ratings. Audit-aware modes use codebase reports or handoff.yaml; standalone option.