review-spec
Skill0 starsChanged 34 days agoMITView on GitHub
Review a PRD, spec, one-pager or set of stories the way a demanding senior PM, engineer and designer would, and return prioritised, specific feedback with suggested rewrites. Use when the user asks to review, critique, sanity-check, tighten or "poke holes in" a product document, or before a spec goes to engineering or leadership.
18 other projects use a version of this.
Adopted by
Most recently taken up by
- fclabs/skillstheir own version, first seen Nov 12, 2026
- 0xdeafcafe/skillstheir own version, first seen Nov 6, 2026
- alo-exp/silver-bullettheir own version, first seen Nov 1, 2026
- holgerleichsenring/specification-first-agentic-developmenttheir own version, first seen Oct 28, 2026
- swm8023/skillstheir own version, first seen Oct 14, 2026
- warpdotdev/oz-for-osstheir own version, first seen Oct 8, 2026
Used with
Projects with it also use create-spec.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
srobinsonSkillcreate-spec
Create a task specification (SPEC.md) through interactive requirements elicitation. Use when helping users define what they want to build before autonomous execution begins.
Also in 3 projects that use this
Other versions18 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 18
- warpdotdev/oz-for-oss★ 313same name
- sd0xdev/sd0x-dev-flow★ 192same name
- sd0xdev/sd0x-harness★ 192same name
- tranhieutt/software_development_department★ 71same name
- holgerleichsenring/specification-first-agentic-development★ 41same name
- Codagent-AI/agent-skills★ 31same name
- alo-exp/silver-bullet★ 7same name
- sergeyklay/.agents★ 6same name
- dekoza/oh-my-slop★ 4same name
- NitinJ/specforge★ 4same name
8 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- acceptance-criteriaSkill
- prioritizeSkill
- slice-storiesSkill
- stakeholder-updateSkill
- write-prdSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.
0
Common
Most of the 17 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 14
Run pm-skills lint <file> (add --type prd, stories, criteria or update if the type is not detected).
- Very specific: line 24
Start from pm-skills template review-spec.
- Distinctive: line 29
Do not rewrite the whole document unless asked.
“Common” is measured: kept by at least 6 in 10 of the 17 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Review a product document
Your job is to find what will cause rework, arguments or a failed launch, while it is still cheap to fix. Be direct and specific. Praise only what the author should keep doing.
CLI: node "${CLAUDE_PLUGIN_ROOT}/cli/pm-skills.mjs" (or ../../cli/pm-skills.mjs relative to this file). Below, pm-skills.
1. Run the mechanical checks first
Run pm-skills lint <file> (add --type prd, stories, criteria or update if the type is not detected). Its findings go in your review as-is, grouped, so you can spend your attention on judgment.
2. Read as three people
- Engineer: Can I estimate this? Which requirement is ambiguous, contradictory or missing an edge case? What dependency or data migration is unmentioned? What will be hard that the doc treats as easy?
- Designer / user: Is the problem real and evidenced? Is the user's journey clear end to end, including errors and empty states? Who is left out?
- Executive: Why now? How will we know it worked: is the success metric the right one, with a baseline and a date? What does it cost, and what are we not doing instead? What is the riskiest assumption, and is there a cheap way to test it first?
3. Write the review
Start from pm-skills template review-spec. Rules:
- Order findings by impact: Blocking (must fix before work starts), Important (fix before launch), Minor (polish).
- Each finding quotes or points to the exact place, explains the consequence if left as is, and proposes a concrete rewrite or question.
- Limit to the 10 findings that matter most. A review with 40 comments gets none of them fixed.
- Do not rewrite the whole document unless asked. Do not change the author's decisions; challenge them with reasons.
- End with the three questions the author should be able to answer before the next review.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

