blueprint
Skill23 starsChanged 2 months agoMITView on GitHub
Use when a written feature spec needs its technical plan: \"plan this spec\", \"how should we build this spec?\", \"make the technical plan for this feature\", a spec in specs/ with no plan.md yet, or a plan that a spec or constitution change has made stale. Not for a feature with no spec (that's `specify`, or `build` for a small one), not for one design question outside any feature (that's the `architect` agent).
66 other projects use a version of this.
Adopted by
Most recently taken up by
- Rojios/everything-claude-codetheir own version, first seen Nov 8, 2026
- sanjaibalajee/skillstheir own version, first seen Nov 8, 2026
- DevvGwardo/blueprint-skilltheir own version, first seen Nov 8, 2026
- dimsedra/skillstheir own version, first seen Nov 8, 2026
- smilewilson1999/blueprinttheir own version, first seen Nov 8, 2026
- pankaj-mahaur/Agent-Daemontheir own version, first seen Nov 6, 2026
Used with
Projects with it also use align, execute and 5 more.
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.
marylinSkillalign
Audit project structure, docs health, and lifecycle compliance. Use when a project feels disorganized, after major refactors, before onboarding a contributor, or periodically as a health check.
Also in 3 projects that use this
sahinSkillexecute
Also in 5 projects that use this
LangChainSkilldocker-patterns
Best practices for Docker containerization and multi-stage builds
Also in 3 projects that use this
lukeleekrSkillscout
GitHub 레포 스카우터. 내 Claude Code + Codex App 환경을 분석하여 필요한 GitHub 레포를 자동 추천하거나, 키워드로 GitHub을 검색합니다. "scout", "스카우트", "깃헙 탐방", "레포 추천", "repo search" 등의 키워드에 반응합니다.
Also in 3 projects that use this
HABUBUSSSkilleval-harness
Also in 3 projects that use this
renatorozasSkillgrill
Adversarial code review. Don't let me ship until the changes pass your scrutiny.
Also in 3 projects that use this
Other versions66 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 66
- WordPress/agent-skills★ 2.3ksame name
- hyperlink-academy/leaflet★ 527same name
- astronomer/agents★ 459same name
- loulanyue/awesome-claude-notes★ 272same name
- Joonghyun-Lee-Frieren/oh-my-antigravity★ 214same name
- Joonghyun-Lee-Frieren/oh-my-gemini-cli★ 214same name
- team-attention/hoyeon★ 173same name
- Fmarzochi/EGC★ 61same name
- manusco/resonance★ 38same name
- smilewilson1999/blueprint★ 27same name
56 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 repository34 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- analyzeSkill
- assessSkill
- breakdownSkill
- buildSkill
- checklistSkill
- clarifySkill
- constitutionSkill
- convergeSkill
- debugSkill
- deepenSkill
- deprecateSkill
- double-checkSkill
- explainSkill
- handoffSkill
- helpSkill
- initSkill
- interviewSkill
- learnSkill
- levelupSkill
- lintSkill
- living-docsSkill
- performanceSkill
- personaSkill
- promptSkill
- prototypeSkill
- qaSkill
- quarantineSkill
- reportSkill
- roadmapSkill
- routeSkill
- signatureSkill
- specifySkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 0 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 65 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
Refuse to plan over open questions: If [NEEDS CLARIFICATION] markers remain, run interview first.
- Very specific: line 22
Write plan.md from specs/.templates/plan.md if present, else template.md.
- Very specific: line 48
The plan is not the code: No function bodies, no full migrations, no test suites.
“Common” is measured: kept by at least 6 in 10 of the 65 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
MasterMind: Blueprint
The plan turns the spec's what into a how that can be checked against the constitution before a
line is written. Layout and IDs: ~/.mastermind/engineering/core/spec-driven.md. Input: $ARGUMENTS.
Before planning
- Find the feature folder and read
spec.mdin full, plusspecs/constitution.md,specs/tech.md, and the active field'sstack-defaults.mdwhen a field pack exists. - Refuse to plan over open questions. If
[NEEDS CLARIFICATION]markers remain, runinterviewfirst. When the user insists on skipping, proceed and record the open markers as risks inplan.md. - Run the constitution gate. For each principle, write
pass,n/aorviolatesin the plan's gate table. A violation stops the plan until the user accepts it in the complexity table.
Plan
Write plan.md from specs/.templates/plan.md if present, else template.md. Then:
- Technical context. Language, dependencies, storage, testing, platform, performance and scale
targets. Each value comes from
tech.md, the code, or a decision you make in step 2. - Research the unknowns. One entry in
research.mdper real decision: chosen, why, rejected and why. Read the primary source for anything version-sensitive (learn). Library choices with real adoption risk go to thetech-scoutagent. - Design. Dispatch the
architectagent with the spec, the constitution andtech.md. Its output becomes the plan's design section. Keep the boundaries, data model, contracts and the reason for each. Where no isolated agent exists, followarchitect's method yourself and say so in the plan. - Write what the design implies, each only when it has content:
data-model.md: entities with fields, validation rules taken from the spec, relationships, state changes.contracts/: every interface another system or user depends on, in the form the project already uses, or a plain table when it has none.quickstart.md: the runnable steps that prove the feature works end to end, with expected output. A short script is allowed when steps alone cannot drive it.
- Name the real paths. The structure section lists the actual directories and files to create or change in this repository. Never a generic layout.
- Name what the spec implies but never says. A reasonable user expects some inputs handled that no
requirement and no edge case names. List up to five of them, such as an empty name, a double submit or a
timezone at midnight. Add each to the plan's risks and give it a task in
breakdown. A spec's silence is not permission to crash. - Re-run the gate against the finished design. Set the spec's status to
planned.
Gotchas
- The plan is not the code. No function bodies, no full migrations, no test suites. Signatures and
shapes only; the code is
build's job. A plan longer than the spec it serves is a sign it wrote code. - Failure is not cancellation. In a state model, a failed attempt, a retry and a cancelled one are different states, and every retry names the state it re-enters.
- Every decision traces to a requirement. A dependency, service or layer nobody asked for goes in the complexity table with the simpler option it beat, or it goes.
- Match the repository. A plan that introduces a second ORM, router or state library has to justify it against the one already there.
- A requirement the plan cannot meet is a finding, not a footnote. Report it to the user before
breakdownruns.
Report
The files written, the decisions with one-line reasons, the gate result before and after design, any
accepted violation, and the next step: breakdown, or checklist first for a sensitive feature.
Then run mastermind next and end with the step it names, so the user sees where the feature stands.
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.

