10x-plan-review

Review implementation plans for feasibility, consistency, and architectural fitness before coding.

Updated May 20, 2026
One-click install
npx skills add https://github.com/ZawilecxD/trAInR --skill 10x-plan-review-zawilecxd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: 10x-plan-review
Source: https://github.com/ZawilecxD/trAInR/tree/main/.cursor/skills/10x-plan-review
Command: npx skills add https://github.com/ZawilecxD/trAInR --skill 10x-plan-review-zawilecxd

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Flawed implementation plans cause wasted engineering hours when problems surface mid-build. This Skill catches substance problems in a plan — contradictions, missing phases, ungrounded file paths, architectural mismatches — before any code is written. ## Core Features & Use Cases - Multi-dimensional analysis: Scores plans across End-State Alignment, Lean Execution, Architectural Fitness, Blind Spots, and Plan Completeness with PASS/WARNING/FAIL verdicts. - Grounding and codebase verification: Checks that referenced file paths and symbols actually exist, and launches a sub-agent to verify the riskiest claims against real code with file:line evidence. - Interactive triage workflow: Presents findings with severity and impact ratings, offers fix options with tradeoffs, applies accepted fixes directly to the plan, and supports saving/resuming review reports. - Use Case: After drafting an implementation plan with /10x-plan, run this Skill to validate the plan against the actual codebase, get a SOUND/REVISE/RETHINK verdict, and triage each finding before starting implementation. ## Quick Start Ask the assistant to review the plan at context/changes/my-change/plan.md and triage the findings.

Frequently Asked Questions about 10x-plan-review

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

FAQPage Schema
How do I review an implementation plan before coding?

Point the Skill at a plan markdown file or a change ID containing plan.md. It scans for contradictions, promise gaps, and contract breaks, verifies referenced paths and symbols against the codebase, then reports findings with severity and fix options.

What does the plan review check for?

It evaluates five dimensions: End-State Alignment, Lean Execution, Architectural Fitness, Blind Spots, and Plan Completeness. It also runs grounding checks on file paths and symbols, plus a deep sub-agent verification of the riskiest claims.

Can I resume a saved plan review later?

Yes. Reports saved with the PLAN-REVIEW-REPORT marker can be passed back as an argument to resume triage. The Skill parses findings with Decision: PENDING and continues the interactive triage loop from where you stopped.

What is the difference between deep mode and quick mode?

Quick mode (--quick flag) skips Step 3 codebase verification, so findings rely only on document analysis and grounding checks. Deep mode launches a sub-agent to verify risky claims, sweep blast radius, and check for pattern proliferation.

When should I not run a plan review?

Archived plans under context/archive/ are refused — the Skill prints a message and stops. It also will not rewrite your plan outside of triage; it only analyzes and reports unless you approve a fix during triage.