blueprint-plan-drafter
Skill5 starsChanged 23 days agoMITView on GitHub
Draft detailed coding plans for cross-module coordination, shared contracts, migrations, compatibility risks, or coordinated multi-agent implementation and review. Use modest-plan-drafter for concise, bounded coding or non-coding plans.
In this repository47 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- agents-md-fixerSkill
- arch-docsSkill
- book-to-markdownSkill
- docset-updaterSkill
- humanizerSkill
- news-release-docsSkill
- pdf-guideSkill
- readme-docsSkill
- screenshot-docsSkill
- see-also-docsSkill
- setup-install-usage-docsSkill
- bptools-writer-expertSkill
- color-accessibility-expertSkill
- css-creative-expertSkill
- game-design-expertSkill
- geometry-expertSkill
- glass-expertSkill
- hic-human-interact-expertSkill
- openscad-3d-printing-expertSkill
- podman-expertSkill
- postgresql-expertSkill
- pyside6-engineerSkill
- rust-code-expertSkill
- solid-js-expertSkill
- svg-creator-expertSkill
- typescript-engineerSkill
- ui-ux-engineerSkill
- vision-expertSkill
- wasm-rust-expertSkill
- webwork-writer-expertSkill
- delegate-manager-to-subagentsSkill
- gas-town-workflowSkill
- hang-checkSkill
- html-game-parallel-builderSkill
- parallel-planSkill
- stay-busySkill
- repo-rules-readerSkill
- ideonomy-plainSkill
- ideonomy-richSkill
- make-goalSkill
- modest-plan-drafterSkill
- audit-code-reviewerSkill
- old-python-code-reviewSkill
- prune-testsSkill
- unit-test-starterSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
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
- Distinctive: line 20
Read the target repository's operative agent/rule authority, or reuse a valid same-session rules receipt while its scope and sources remain …
- Distinctive: line 29
Read testing, language, UI, database, documentation, runtime, deployment, security, or other specialized guidance only when the proposed …
- Very specific: line 31
Read refactor_progress.md or relevant active plans when they exist and affect coordination.
- Very specific: line 32
Read references/PLAN_TEMPLATE_EXAMPLE.md only when an archetype example will clarify the plan shape.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Blueprint plan drafter
Purpose and routing
Draft planning and progress documentation for detailed technical decisions and coordination. Use
modest-plan-drafter when a concise approach and completion checks adequately describe the work.
Honor explicit skill choices; if a requested Modest plan leaves important technical decisions
unresolved, explain the need for Blueprint or expand the relevant detail. Let decision complexity
determine detail, rather than document length.
Authority model
Always-read authorities
- Read the target repository's operative agent/rule authority, or reuse a valid same-session rules receipt while its scope and sources remain unchanged.
- Read
references/PLAN_HEADINGS.md,references/PLAN_TEMPLATE_BLANK.md,references/plan_quality_standard.md, andreferences/DEFINITIONS.md.
Task-dependent authorities
- Read testing, language, UI, database, documentation, runtime, deployment, security, or other specialized guidance only when the proposed work touches that concern.
- Read
refactor_progress.mdor relevant active plans when they exist and affect coordination. - Read
references/PLAN_TEMPLATE_EXAMPLE.mdonly when an archetype example will clarify the plan shape.
Repository evidence
Inspect the architecture, code, tests, documentation, current behavior, and recent decisions needed to establish the actual implementation boundary. Treat older plans as evidence rather than current implementation authority.
For broad code investigation, read GRAPHIFY_GUIDE.md when a Graphify map or repository wrapper is available. Otherwise use targeted searches and source inspection.
Workflow
- Draft from the blank template using the canonical core: Context, Objectives, Design philosophy, Scope, and Non-goals. Add only applicable sections, following the headings reference.
- State assumptions, constraints, component responsibilities, and authorization boundaries. Give file-based work a concise File scope under Architecture boundaries and ownership, or Files to modify for the small archetype: expected paths and changes, including relevant tests, docs, and generated outputs with their canonical sources. Refine this boundary as evidence emerges.
- Define milestone deliverables, meaningful dependencies, entry and exit criteria, and a brief
Parallel-plan ready: yes/noreason. Workstreams may name independent areas; detailed task assignments, package IDs, and agent counts belong to execution. - Separate observations, hypotheses, and decisions. Resolve uncertain methods through bounded comparisons with success measures and correction paths. Keep one abstraction level per plan: stabilization, technical redesign, or program coordination. Apply the quality standard's scrap-vs-fix criteria when evidence challenges the design.
- Select verification using the quality standard's KISS, robustness, configuration, and test-retention criteria. Ground requirements in actual needs and give new blocking gates a failure response.
- Cover material risks with impact, trigger, owner, and mitigation, plus the documentation, generation, rollout, and release work required for closure.
Handoff
Planning ends with the plan artifact, before production code or tests. For a requested next stage,
use make-goal to distill an outcome-only goal, or read
EXECUTION_RESOURCES.md for execution and review routing.
Completion criteria
Finish when the plan is internally consistent, its outcomes and boundaries are clear, and every execution path has appropriate completion evidence. Resolve blocking choices or define an evidence-led investigation before dependent work; remaining follow-up questions are non-blocking.
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.

