gstack-openclaw-ceo-review

Reviews implementation plans across eleven sections with scope-mode selection and error-path analysis.

1|Updated Jul 23, 2026
One-click install
npx skills add https://github.com/sanjanb/my-agent-harness --skill gstack-openclaw-ceo-review-sanjanb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gstack-openclaw-ceo-review
Source: https://github.com/sanjanb/my-agent-harness/tree/main/skills/openclaw/skills/gstack-openclaw-ceo-review
Command: npx skills add https://github.com/sanjanb/my-agent-harness --skill gstack-openclaw-ceo-review-sanjanb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Plans often ship with hidden failure modes, vague error handling, and unexamined scope. This Skill applies a rigorous CEO-style review to any plan before implementation, catching landmines early and forcing explicit scope decisions. ## Core Features & Use Cases - Scope Mode Selection: Choose between scope expansion, selective expansion, hold scope, or scope reduction, with every scope change requiring explicit user opt-in. - Eleven-Section Review: Systematically evaluates architecture, error handling, security, data flows, code quality, tests, observability, database design, API contracts, performance, and UX. - Structured Summary Output: Produces a CEO review summary with strongest challenges, recommended path, accepted scope, and deferred items saved to memory. - Use Case: Before implementing a new feature touching multiple services, run this review to map every error path, challenge the premise, compare 2-3 implementation alternatives, and confirm the plan is bulletproof. ## Quick Start Ask the agent to run a CEO review on your implementation plan and challenge its scope before writing any code.

Frequently Asked Questions about gstack-openclaw-ceo-review

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

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

Run a CEO plan review that challenges the premise, compares 2-3 implementation alternatives, and evaluates eleven sections including architecture, error handling, security, and tests. No code changes are made during the review.

What review modes does this plan review support?

Four modes are supported: scope expansion, selective expansion, hold scope, and scope reduction. Defaults depend on context, such as expansion for greenfield features and hold scope for bug fixes or refactors.

Can the review change my plan's scope automatically?

No. Every scope change is an explicit opt-in decision made by the user. Expansion ideas are presented individually so you can accept or reject each one separately.

Does the CEO review skill write or modify code?

No. The skill only reviews plans and explicitly forbids code changes or implementation. Its output is a structured review summary with challenges, recommendations, and accepted scope.

What sections does the plan review cover?

Eleven sections: architecture, error and rescue mapping, security and threat model, data flow edge cases, code quality, tests, observability, database and state, API design, performance, and UX when the plan touches UI.