gepetto

Creates detailed, sectionized implementation plans through research, stakeholder interviews, and multi-LLM review. Use when planning features that need thorough pre-implementation analysis.

2.40x

Quality

76%

Does it follow best practices?

Impact

96%

2.40x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./dist/plugins/gepetto/skills/gepetto/SKILL.md

Evaluation results

100%

49%

Break Down an Implementation Plan into Workable Sections

Section manifest format and structure

Criteria

Without context

With context

Manifest block at top

100%

Two-digit section numbers

50%

100%

Lowercase hyphenated names

50%

100%

Sequential numbering

100%

Dependency graph present

30%

100%

Execution order present

50%

100%

Section summaries present

50%

100%

Section files created

100%

Background section in files

100%

Requirements section in files

100%

Dependencies section in files

66%

100%

Acceptance criteria in files

100%

Files to create listed

100%

90%

75%

Automate Implementation Plan Review with External AI Reviewers

External review CLI commands

Criteria

Without context

With context

Gemini model flag

100%

Gemini approval flag

100%

Codex exec subcommand

100%

Codex model flag

100%

Codex sandbox flag

100%

Codex skip-git flag

100%

Codex full-auto flag

100%

Codex stderr suppressed

100%

Parallel execution

Gemini output file

50%

100%

Codex output file

50%

100%

Failure handling

87%

100%

44%

Generate Autonomous Implementation Handoff Files

Execution files generation

Criteria

Without context

With context

ralph-loop file created

100%

Sections embedded in ralph-loop

100%

Mission statement present

100%

Index embedded in ralph-loop

62%

100%

Completion signal present

100%

Execution rules in ralph-loop

100%

Ralphy PRD file created

100%

PRD references not embeds

60%

100%

Checkbox task list format

100%

Ralphy usage instructions

100%

All sections covered in PRD

100%

Repository: softaworks/agent-toolkit
Commit: 3027f20

Evaluated: 30 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Break Down an Implementation Plan into Workable Sections Automate Implementation Plan Review with External AI Reviewers Generate Autonomous Implementation Handoff Files

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.