skills.rest
srnnkls
srnnklsCOMMUNITY
Explore all skills by this author →
GitHub

spec-review

Coordinate parallel Claude and OpenCode spec reviews with YAML reports.

1|Updated Jan 12, 2026
Search Tags:#automation#workflow#multi-agent#yaml#review#spec
One-click install
npx skills add https://github.com/srnnkls/tropos --skill spec-review-srnnkls
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: spec-review
Source: https://github.com/srnnkls/tropos/tree/main/skills/spec-review
Command: npx skills add https://github.com/srnnkls/tropos --skill spec-review-srnnkls

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Spec reviews often suffer from fragmented feedback and misaligned perspectives. This skill coordinates multiple reviewer personas to validate specs in parallel, reducing cycle time and increasing confidence before implementation.

Core Features & Use Cases

  • Parallel review orchestration with Claude and OpenCode reviewers to surface issues quickly.
  • YAML-based reviewer reports and synthesized gates for clear decision-making.
  • Reference materials and playbooks for consistent evaluation and edge-case handling.
  • Standalone or integrated workflow with commands like /spec.review, /spec.promote, and related tasks.

Quick Start

Run a parallel multi-agent spec review with /spec.review <spec-name> to begin.

Frequently Asked Questions about spec-review

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

FAQPage Schema
How do I run a multi-agent spec review to validate my project requirements?▼

A multi-agent spec review dispatches Claude and OpenCode reviewers in parallel to validate your spec, generating YAML-based reports that aggregate gates, issues, and clarifying questions for improved spec quality.

What is the best way to automate parallel spec reviews for pre-implementation checks?▼

Automating parallel spec reviews involves orchestrating multiple reviewer personas simultaneously to surface issues quickly, reducing cycle time and increasing confidence before implementation begins.

Can I use YAML-based reports to synthesize fragmented spec review feedback?▼

YAML-based reviewer reports provide structured feedback from parallel spec reviews, allowing a synthesis workflow to aggregate gates, issues, and clarifying questions into clear decision-making outputs.

How do I coordinate multiple reviewer personas to reduce spec review cycle time?▼

Coordinating multiple reviewer personas involves dispatching parallel reviews across spec creation, revision, and pre-implementation tasks, using reference playbooks for consistent evaluation and edge-case handling.

Does the multi-agent spec review workflow require any external dependencies?▼

The multi-agent spec review workflow operates standalone or integrated with commands like /spec.review and /spec.promote, requiring no external dependencies to orchestrate parallel Claude and OpenCode reviewers.

When should I avoid using parallel multi-agent spec reviews?▼

Parallel multi-agent spec reviews are not ideal for simple, single-perspective checks where fragmented feedback is not a concern, as the overhead of orchestrating multiple reviewer personas outweighs the synthesis benefits.

Related Skills

Explore More Software Engineering Skills→
Lands-1203Lands-1203
40

f2s-doc-final

Convert PDF and Markdown documents into a standardized final specification format.

Community
Intermediate
spring-ai-alibabaspring-ai-alibaba
2.8k

web-research

Plan and execute web research with subagent delegation and synthesized reports.

Official
Intermediate
TymorIbrahimTymorIbrahim

dmux-workflows

Manage parallel AI agent sessions with dmux across Claude Code, Codex, and OpenCode.

Community
Intermediate
matheushenriquefsmatheushenriquefs

mentat-implement

Execute mentat plans slice-by-slice with TDD and gate checks.

Community
Advanced