planning-reviews

Run planning-phase reviews and log Accept/Reject/Defer dispositions into the plan artifact.

1|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/Atlas-Memory-Framework/AtlasMemory-Tools --skill planning-reviews
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: planning-reviews
Source: https://github.com/Atlas-Memory-Framework/AtlasMemory-Tools/tree/main/.cursor/skills/planning-reviews
Command: npx skills add https://github.com/Atlas-Memory-Framework/AtlasMemory-Tools --skill planning-reviews

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill ensures planning-phase plans are understandable, structurally complete, and safely positioned for implementation by running required review passes and recording findings with dispositions in the plan artifact.

Core Features & Use Cases

  • Planning review orchestration: Executes the required planning reviews (zero-context, implementer readiness, expert technical when triggered, security/privacy, human readability, and automation readiness when applicable) and merges the results into the plan’s ## Planning Reviews section.
  • Evidence-grounded security/privacy checks: Performs a security/privacy review grounded in the current plan state and requires explicit evidence hooks for trust-boundary claims, emitting findings that drive remediation when evidence is missing.
  • Deterministic disposition workflow: Produces stable, review-type-specific findings with F-xxx ids and exhaustive dispositions (Accept/Reject/Defer), including DR references and patch targets.
  • Safety guardrails for human agency: Supports auto-remediation only for clarity/structure issues that don’t change decisions, while requiring explicit human A/R/D for policy, trust boundaries, unknowns, or decision-affecting items.
  • Required review schemas: Uses the mandated review schemas so downstream validators and agents can parse findings and dispositions reliably.

Quick Start

Use the planning-reviews skill during the Reviews stage (after Implementation but before plan approval) to refresh planning review findings and write dispositions into the current plan artifact for the next approval loop.

Frequently Asked Questions about planning-reviews

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

FAQPage Schema
How do I validate a plan for approval readiness before implementation?

Plan validation for approval readiness is executed by running required planning reviews and logging human dispositions into the plan artifact. This verifies readability, implementer readiness, technical risk, and security/privacy.

What is the disposition workflow for logging planning review findings?

The disposition workflow produces stable F-xxx finding IDs with exhaustive Accept, Reject, or Defer dispositions. It includes DR references and patch targets, merging results into the plan's Planning Reviews section.

How do security and privacy reviews handle trust-boundary claims in a plan?

Security and privacy reviews require explicit evidence hooks for trust-boundary claims. Missing evidence emits findings that drive remediation, ensuring assertions are grounded in the current plan state.

Can automated remediation be used for all planning review findings?

Automated remediation only applies to clarity or structure issues that do not alter decisions. Policy, trust boundaries, unknowns, and decision-affecting items require explicit human A/R/D dispositions to preserve agency.

When should I run planning reviews during the development lifecycle?

Run planning reviews during the Reviews stage, after implementation-stage content exists but before plan approval. This refreshes planning review findings and writes dispositions for the next approval loop.

Do planning review outputs follow a specific schema for downstream automation?

Planning review outputs use mandated review schemas so downstream validators and agents can reliably parse findings and dispositions. The post-remediation output format preserves authority boundaries for automated processing.