Summarize AI-generated branches, PRs, diffs, and design docs into structured digests with file breakdown, code walkthrough, and key concepts. Supports audience targeting (-f manager/pm/designer/non-technical/any string) and export (-e md for markdown with Mermaid, -e html for a self-contained HTML explainer). Use when…
Explain what a file, module, or symbol is, how it works end-to-end, why it was designed this way, or where a bug comes from — code reading plus git archaeology (log, blame, PR history), output as a self-contained HTML explainer with inline SVG diagrams. Use for state questions about existing code: a path, a symbol, a…
Generate release notes from a commit range with developer and user-facing formats. Use when creating changelogs, release notes, or summarizing changes between tags.
Synthesize git activity across multiple projects over a time window into themed recall notes. Clusters commits into themes (features, fixes, refactors, infra, learning surface, tickets, side quests), auto-hides empty sections, and surfaces work that PM tools don't track. Use when the user asks 'what did I do last…
Frame a feature, bug, or design question before building: scout the codebase, surface assumptions or interview, compare options, and converge on an explicit decision. Use when an idea needs sharpening before a spec or plan exists. Read-only — never implements.
Commit staged changes with auto-generated Conventional Commits messages. Use when committing changes, finishing work in submodules, or creating structured commit messages.
Show per-submodule git status with staged/unstaged/untracked counts and commit-readiness indicator. Use when checking submodule state or preparing to commit.
Review implementation against a spec using git diffs to catch mismatches, missing work, and bugs. Supports iterative review+fix with --fix flag and parallel multi-angle review. Use after implementing a feature to verify spec coverage.
Dual-engine spec review — Claude and Codex review the same change in parallel, then a Workflow loops fixes until BOTH engines are MEDIUM-clean. Use when a spec is high-stakes and you want two different models to cross-check for blind spots before implementation. Requires the openai-codex plugin for the Codex engine.
Review a spec, proposal, or design before implementation to catch gaps, risks, and ambiguities. Runs multi-angle parallel review by default; scans codebase for context first. Use --fix for iterative review+fix loop (agent escalates design-judgment issues to you), --angles to narrow the angle set.