Product Spec Reviewer — Testing & QA agent for Claude Code
Reviews SPEC/NN before PR 1 of a milestone, and SPEC/00 as a whole at adoption.
How to install Product Spec Reviewer
Installs to ~/.claude/agents/andaro74-extrasforge-product-spec-reviewer.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/andaro74/extrasforge/HEAD/.claude/agents/product-spec-reviewer.md -o ~/.claude/agents/andaro74-extrasforge-product-spec-reviewer.md Restart Claude Code, or start a new session, for it to be picked up.
What Product Spec Reviewer does
name: product-spec-reviewer description: Reviews SPEC/NN before PR 1 of a milestone, and SPEC/00 as a whole at adoption. Checks the false state is named, the plain sentence is plain, the cut list is ordered. Its report is pasted into milestones/MNN/feasibility.md (at adoption, milestones/adoption/spec00-review.md). A report, never a ruling. seat: Product tools: Read, Grep, Glob
You review one milestone's spec before its PR 1 is written, or SPEC/00 as a whole at adoption. You serve the P
Alternatives in Testing & QA
- Fable Verifier — Cold verifier for finished fable deliverables 837 ★
- Confluence Searcher — Searches Confluence and related tickets for product, architecture, rollout, and test-data context 548 ★
- Propagate — Generate tests from Allium specifications 474 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Evaluate
Evidence-based auditor in a fresh context. Verifies every claim in one named artifact (a spec cut, a board's c
Spectastic Classifier
Classifies one pasted item to the spectastic lifecycle layer that owns its fix, for /spectastic.triage list-in
Engine Implementer
C++/pyext implementation agent for the orkid engine. Use to implement ONE precisely-scoped slice of engine wor
Plan Judge Agent
Judges a whole set of step plans as a set — does it actually reach a finished product? Checks coverage against
Module Coder
Use this agent when a feature has been decomposed by feature-decomposer into a product spec with feature list,
Mx Test Builder
Use this agent to write Playwright end-to-end tests for a built product. It takes the feature list and test sc
Related Skills
Sliced Review
Claude Code skill that has other AI models (GPT, Grok, GLM, and more) review your whole codebase for bugs. It
Deep App Plan
Builds a whole application from nothing. Frames product, settles platform decisions, scaffolds stack, deploys
Batch Evaluate
Batch evaluate PENDING pattern adoptions from the adoption log. Reviews evidence, checks artifact existence, a