bpe

Plugin bpe from marketplace bpe in MasonEgger/bpe-claude-code-plugin. Commands read from the repo's manifest — 3 days ago.

Skills (13)

• apply-review

  Load saved review feedback from /bpe:review and apply changes to the reviewed artifact

• brainstorm

  Iterative Q&A to develop a thorough project specification (spec.md), or to merge the next phase into an existing one

• commit-message

  Generate a commit message explaining what was changed and write it to commit-msg.md

• execute-plan

  Implement the next unchecked step from plan.md by executing its sub-steps as written (RED-GREEN-REFACTOR for Feature steps; Scope/Tooling/Do/Verify/Document for Task steps)

• gh-issue

  Fetch a GitHub issue and route to brainstorm or plan based on detail level

• goal

  Autonomous-mode BPE run via /goal. Modes are full (default) | section <name> | step. Pre-flights branch safety (refuses on main, requires a clean tree), resolves the project verification command (test-runner autodetect, spec.md fallback, or ask), builds a verifiable completion condition, and writes

• handoff

  Create, continue, or close handoff documents for fresh-agent baton passes

• lessons

  View, search, and manage accumulated lessons from .ai-sessions/lessons.md

• plan

  Transform spec.md into an implementation roadmap (plan.md + todo.md) of TDD Feature steps and non-TDD Task steps

• retrofit

  Retrofit a BPE-compatible spec.md onto an existing project that lacks one, or resync an existing spec.md to the code, migrating a pre-0.7 spec to the canonical section order along the way. Reads repo state and runs a shortened Q&A focused on gaps.

• review

  Render spec.md, plan.md, or todo.md as a fine-grained, decision-by-decision review page in your browser, then write the feedback as JSON for /bpe:apply-review to consume.

• session-summary

  Generate session recap in .ai-sessions/ and capture lessons learned

• wtf-wid

  WTF was I doing? A tight, fits-on-screen recap of the current session so you can re-enter in under 10 seconds

Subagents (3)

• cheap-research

  Fast, cheap external research subagent for tool discovery, docs lookup, quick fact-checks. Dispatched by /bpe:plan, /bpe:brainstorm, /bpe:retrofit when they need external info.

• step-executor

  Executes ONE BPE step as the worker for /bpe:goal autonomous mode. Operates in one of three modes set by the orchestrator: `implement` (execute the plan step's sub-steps, no commit), `fix` (apply validator findings, no commit), `finalize` (session summary, single commit, push). Backwards-compatible

• validator

  Generic read-only QA agent dispatched by /bpe:goal between executor mode=implement and mode=finalize. Reads the uncommitted diff, consults MCP servers and skills passed in by the orchestrator (from plan.md per-section declarations), runs any linters declared in the section's Tools block, and returns

Similar plugins