skills.rest
sfc-gh-myoung
sfc-gh-myoungCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

plan-reviewer

Analyze LLM-generated plans for autonomous execution across three review modes.

7|1|Updated Oct 31, 2025
Search Tags:#automation#consistency#rubric#plan-reviewer#executability#llm-plans#delta-review
One-click install
npx skills add https://github.com/sfc-gh-myoung/ai_coding_rules --skill plan-reviewer-sfc-gh-myoung
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: plan-reviewer
Source: https://github.com/sfc-gh-myoung/ai_coding_rules/tree/main/skills/plan-reviewer
Command: npx skills add https://github.com/sfc-gh-myoung/ai_coding_rules --skill plan-reviewer-sfc-gh-myoung

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Plan-reviewer evaluates LLM-generated plans to ensure autonomous agents can execute them without needing human guidance.

Core Features & Use Cases

  • FULL mode: single-plan evaluation with eight dimensions and a clear verdict.
  • COMPARISON mode: rank multiple plans and declare a winner.
  • META-REVIEW: analyze cross-review consistency and calibration.
  • DELTA mode (optional): track improvements and regressions across iterations.

Quick Start

Load the plan-reviewer skill and run a FULL mode review on a single plan file to validate executability and completeness.

Frequently Asked Questions about plan-reviewer

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

FAQPage Schema
How do I evaluate if an LLM-generated plan is executable by autonomous agents?▼

You evaluate LLM-generated plans by running a FULL mode review that scores executability across eight dimensions, producing a clear verdict on whether autonomous agents can execute the plan without human intervention.

What is the best way to compare multiple LLM plans and rank them by executability?▼

The best way to compare multiple LLM plans is using a COMPARISON mode review, which evaluates plans against a rubric, ranks them by executability, and declares a clear winner for autonomous agent execution.

How do I track improvements and regressions across iterative LLM plan versions?▼

You track iterative LLM plan changes by running a DELTA mode review, which analyzes version differences to identify specific improvements and regressions across plan iterations for autonomous execution readiness.

Can I analyze cross-review consistency for LLM plans evaluated by different models?▼

Yes, you can analyze cross-review consistency for LLM plans using a META-REVIEW mode, which examines calibration and consistency across multiple reviews to ensure uniform executability scoring standards.

What inputs are required to review the executability of an LLM-generated plan?▼

Reviewing LLM-generated plan executability requires review_date, review_mode, and model as inputs. The rubric-based scoring then generates a final verdict written to a reviews/ directory.

Related Skills

Explore More Software Engineering Skills→
ExuberantWitnessExuberantWitness
1

flux-verify-paper-plan

Validate JSON research paper plans via schema checks and LLM rubric evaluation.

Community
Intermediate
d-o-hubd-o-hub
11

loop-agent

Orchestrate workflow agents in iterative refinement loops with configurable termination modes.

Community
Advanced
AeyeOpsAeyeOps
3

opus-prompting

Optimize Claude Opus 4.5 prompts for literal instruction-following accuracy.

Official
Intermediate
xurenluxurenlu
1

agentic-eval

Evaluate and iteratively refine AI agent outputs through self-critique.

Community
Advanced