Generate stub-level implementation plans from ADRs and research documents.
npx skills add https://github.com/Epiphytic/brains --skill map-epiphytic
Please help me install this Agent Skill. Skill: map Source: https://github.com/Epiphytic/brains/tree/main/skills/map Command: npx skills add https://github.com/Epiphytic/brains --skill map-epiphytic
Projects often lack a clear, reviewable high-level plan that connects architecture decisions to implementable tasks, which makes multi-agent or multi-phase work costly and error-prone. The map skill converts accepted ADRs and research into a concise, stub-level plan that surfaces dependencies, acceptance criteria, and a user approval gate so teams can proceed to implementation with confidence.
Use the map skill to outline a stub-level implementation plan and create beads tasks for the topic "add a health-check endpoint".
High-intent search queries and answers about installing and using this skill.
To generate an implementation plan from an architecture decision record, you can use a tool that parses accepted ADRs and research documents to produce a concise, stub-level plan. This plan surfaces task dependencies, acceptance criteria, and a user approval gate for confident multi-agent implementation.
A stub-level implementation plan is a high-level document that connects architecture decisions to implementable tasks without detailing specific code. You need one when orchestrating multi-agent or multi-phase work to ensure dependencies and acceptance criteria are reviewable before implementation begins.
Yes, you can use multi-LLM review modes to refine task ordering, sizing, coverage, and phasing. By running subagents or star-chamber reviews in single, parallel, or debate modes, the plan is systematically improved before implementation begins.
You can create tracked tasks and topic branches from an accepted ADR by deriving a topic slug and generating a map document. This process automatically creates beads-labeled tasks with brains prefixes and selects the appropriate task tracker for umbrella secure and nurture tasks.
Yes, this plan generation approach works by intentionally avoiding implementation specifics. It focuses on enumerating high-level tasks, dependencies, and acceptance criteria to provide a reviewable high-level plan that connects architecture decisions to actionable work.
prismaCreate project workspaces with specs, plans, and assets directories.
DicklesworthstoneCoordinate parallel software tasks across multiple agents using Agent Mail and BV prompts.
EpiphyticReview and fix code implementations for correctness and test coverage.
mietheGenerate PRDs and phased implementation plans with subagent assignments.
