skills.rest
kaiohenricunha
kaiohenricunhaCOMMUNITY
Explore all skills by this author →
𝕏@kaiohenricunhaGitHub

plan-grader

Grade implementation plans on a 0-10 scale using a weighted evidence-based rubric.

Updated Apr 14, 2026
Search Tags:#code review#rubric scoring#plan grading#implementation plan review#agentic CLI#dev governance#harness validation
One-click install
npx skills add https://github.com/kaiohenricunha/dotbabel --skill plan-grader
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: plan-grader
Source: https://github.com/kaiohenricunha/dotbabel/tree/main/plugins/dotbabel/templates/claude/skills/plan-grader
Command: npx skills add https://github.com/kaiohenricunha/dotbabel --skill plan-grader

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Manually evaluating agent-generated implementation plans for completeness, safety, and alignment with project rules risks accidental workspace changes and inconsistent scoring. This Skill eliminates that risk with a fully non-mutating, evidence-based grading workflow.

Core Features & Use Cases

  • Weighted Rubric Scoring: Grades plans on 7 evidence-based dimensions (scope clarity, intent alignment, repo grounding, execution order, harness compatibility, validation, and risk handling) for an objective 0-10 score.
  • Multi-Agent Support: Accepts plan files, inline quoted text, or auto-discovers latest plans from Claude Code, Codex, GitHub Copilot, and Gemini CLI.
  • Blind Scoring Mode: Strips agent attribution to eliminate bias when comparing plans from different tools.
  • Use Case: Engineering leads can quickly validate implementation plans from their team's agentic CLIs to catch missing validation, harness violations, or risky changes before coding begins.

Quick Start

Use the plan-grader skill to score your latest Codex implementation plan and receive a list of blocking gaps to address before running the plan.

Frequently Asked Questions about plan-grader

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I grade an implementation plan without modifying my workspace?▼

You can grade an implementation plan safely by using a non-mutating, evidence-based rubric that scores scope clarity, intent alignment, and risk handling without writing any changes to the workspace.

What is the best way to review agentic CLI implementation plans for CI governance?▼

Reviewing agentic CLI implementation plans for CI governance is best done by applying a weighted rubric that validates harness compatibility and outputs structured JSON for automated pipeline integration.

Can I automatically discover and grade plans generated by Claude Code, Codex, GitHub Copilot, and Gemini CLI?▼

Yes, you can auto-discover the latest plans generated by Claude Code, Codex, GitHub Copilot, and Gemini CLI, or you can manually provide plan files and inline quoted text for grading.

Does plan grading support blind scoring to eliminate bias when comparing different agent tools?▼

Blind scoring is supported during plan grading by stripping agent attribution, allowing you to objectively compare implementation plans from different agentic CLIs without tool bias.

How do I validate that an implementation plan aligns with project harness rules?▼

Validating implementation plan alignment with project harness rules involves checking the plan against an evidence-based rubric that flags harness violations, missing validation, and risky changes before coding begins.

Related Skills

Explore More Software Engineering Skills→
wiltaylorwiltaylor

oz-platform

Manage cloud agents via Warp's REST API and CLI.

Community
Advanced
supersuit-techsupersuit-tech
4

fix-ci

Trigger GitHub Actions CI and audit workflows, collect failure logs, and iterate until both pass.

Official
Intermediate
stephendorstephendor
1

assay

Automate viability scoring for Discovery Harness candidates using a structured rubric.

Community
Advanced
Ankh-StudioAnkh-Studio

prompt

Evaluate natural language prompts against a structured rubric with weighted scores.

Official
Intermediate