spec-panel

Coordinate five reviewer agents to evaluate a spec or design document.

Updated Aug 27, 2026

Search Tags:#multi-agent #consensus #peer-review #spec-review #design-analysis #lenses

Ready to run

npx skills add https://github.com/icyrainz/.dotfiles --skill spec-panel

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: spec-panel
Source: https://github.com/icyrainz/.dotfiles/tree/main/claude/skills/spec-panel
Command: npx skills add https://github.com/icyrainz/.dotfiles --skill spec-panel

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinate a multi-perspective spec/design review workflow across five reviewer agents, each embodying a distinct design philosophy, to surface gaps, risks, and clarities before implementation.

Core Features & Use Cases

Parallel review: five reviewer personas evaluate the same spec or design document to ensure coverage of scope, completeness, risk, simplicity, and clarity.
Structured findings: each reviewer returns a concise list of findings with severity and rationale, followed by a consolidated consensus.
Decision-ready output: produces a final verdict plus a tensions section that captures disagreements for resolution.

Quick Start

Invoke the panel on a target spec to generate parallel reviews and a consolidated report ready for stakeholder review.

Frequently Asked Questions about spec-panel

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

FAQPage Schema

How do I coordinate a multi-perspective design review for a specification document?▼

You can conduct peer-review of specs by running five reviewer personas in parallel, each returning a concise list of findings with severity and rationale, followed by a consolidated consensus and a final verdict ready for implementation.

What is the best way to surface gaps and risks in a design document before implementation?▼

Surfacing gaps and risks in a design document is achieved by applying multi-agent consensus, where five reviewer personas evaluate scope, completeness, risk, simplicity, and clarity, capturing any disagreements in a tensions section for stakeholder resolution.

How do I generate a decision-ready verdict from a spec review?▼

Generating a decision-ready verdict involves routing a target spec through a consolidated findings pipeline with per-reviewer prompts, which outputs structured findings, a consensus, and a final verdict ready for stakeholder review.

Can I use multi-agent consensus to evaluate the same spec across different design philosophies?▼

Yes, multi-agent consensus can evaluate the same spec across five reviewer agents embodying distinct design philosophies, applying a consistent prompt format to ensure structured coverage of scope, completeness, risk, simplicity, and clarity.

Do I need any dependencies to run a parallel spec review?▼

No dependencies are required to run a parallel spec review; the workflow operates independently to generate per-reviewer prompts, apply consistent formatting, and produce a final consolidated verdict from the target design document.