Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate a structured implementation plan from a draft document. Validates input, checks relevance, analyzes for issues, and generates a complete plan.md with acceptance criteria.
.claude/skills/humanize-gen-plan/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | — | — |
| case-04 | ✗→✓ | ▲ Improved | — | — |
| case-11 | ✗→✓ | ▲ Improved | — | — |
| case-10 | ✗→✓ | ▲ Improved | — | — |
| case-05 | ✗→✓ | ▲ Improved | — | — |
Transforms a rough draft document into a well-structured implementation plan with clear goals, acceptance criteria (AC-X format), path boundaries, and feasibility suggestions.
The installer hydrates this skill with an absolute runtime root path:
bash{{HUMANIZE_RUNTIME_ROOT}}
mermaidflowchart TD BEGIN([BEGIN]) --> VALIDATE[Validate input/output paths<br/>Run: {{HUMANIZE_RUNTIME_ROOT}}/scripts/validate-gen-plan-io.sh --input <draft> --output <plan>] VALIDATE --> CHECK{Validation passed?} CHECK -->|No| REPORT_ERROR[Report validation error<br/>Stop] REPORT_ERROR --> END_FAIL([END]) CHECK -->|Yes| READ_DRAFT[Read input draft file] READ_DRAFT --> CHECK_RELEVANCE{Is draft relevant to<br/>this repository?} CHECK_RELEVANCE -->|No| REPORT_IRRELEVANT[Report: Draft not related to repo<br/>Stop] REPORT_IRRELEVANT --> END_FAIL CHECK_RELEVANCE -->|Yes| ANALYZE[Analyze draft for:<br/>- Clarity<br/>- Consistency<br/>- Completeness<br/>- Functionality] ANALYZE --> HAS_ISSUES{Issues found?} HAS_ISSUES -->|Yes| RESOLVE[Engage user to resolve issues<br/>via AskUserQuestion] RESOLVE --> ANALYZE HAS_ISSUES -->|No| CHECK_METRICS{Has quantitative<br/>metrics?} CHECK_METRICS -->|Yes| CONFIRM_METRICS[Confirm metrics with user:<br/>Hard requirement or trend?] CONFIRM_METRICS --> GEN_PLAN CHECK_METRICS -->|No| GEN_PLAN[Generate structured plan:<br/>- Goal Description<br/>- Acceptance Criteria with TDD tests<br/>- Path Boundaries<br/>- Feasibility Hints<br/>- Dependencies & Milestones] GEN_PLAN --> WRITE[Write plan to output file<br/>using Edit tool to preserve draft] WRITE --> REVIEW[Review complete plan<br/>Check for inconsistencies] REVIEW --> INCONSISTENT{Inconsistencies?} INCONSISTENT -->|Yes| FIX[Fix inconsistencies] FIX --> REVIEW INCONSISTENT -->|No| CHECK_LANG{Multiple languages?} CHECK_LANG -->|Yes| UNIFY[Ask user to unify language] UNIFY --> REPORT_SUCCESS CHECK_LANG -->|No| REPORT_SUCCESS[Report success:<br/>- Plan path<br/>- AC count<br/>- Language unified?] REPORT_SUCCESS --> END_SUCCESS([END])
Required Arguments:
--input <path/to/draft.md> - The draft document--output <path/to/plan.md> - Where to write the planThe generated plan includes:
markdown# Plan Title ## Goal Description Clear description of what needs to be accomplished ## Acceptance Criteria - AC-1: First criterion - Positive Tests (expected to PASS): - Test case that should succeed - Negative Tests (expected to FAIL): - Test case that should fail ## Path Boundaries ### Upper Bound (Maximum Scope) Most comprehensive acceptable implementation ### Lower Bound (Minimum Scope) Minimum viable implementation ### Allowed Choices - Can use: allowed technologies - Cannot use: prohibited technologies ## Dependencies and Sequence ### Milestones 1. Milestone 1: Description - Phase A: ... - Phase B: ... ## Implementation Notes - Code should NOT contain plan terminology
| Exit Code | Meaning | |-----------|---------| | 0 | Success - continue | | 1 | Input file not found | | 2 | Input file is empty | | 3 | Output directory does not exist | | 4 | Output file already exists | | 5 | No write permission | | 6 | Invalid arguments | | 7 | Plan template file not found |
bash# Start the flow /flow:humanize-gen-plan # The flow will ask for: # - Input draft file path # - Output plan file path
Or with the skill only (no auto-execution):
bash/skill:humanize-gen-plan
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-20 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-22 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-21 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-19 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-15 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-17 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-04 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-11 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-13 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-10 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-05 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-01 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-18 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-06 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-09 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-08 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-03 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-07 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-12 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-14 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-16 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
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 19 counted toward the lift figure. The other 3 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 +45 percentage points is the difference between those two pass rates over the 19 comparable cases.
The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.
Other measured skills in the registry, with their headline benchmark lift.