Planning

Use when asked to "investigate", "analyze", "what would be involved", "scope this work", "research", "discover", "look into", "audit docs", or "research [tool/library]". Self-classifies the discovery type and dispatches to a playbook. Callers must not pass a subtype.

Bumps version files (package.json, plugin.json, pyproject.toml, Cargo.toml, VERSION, etc.) based on a diff. Project-type aware — Claude Code plugin markdown under agents/, skills/, commands/, hooks/, and .claude-plugin/ is treated as runtime configuration, not prose docs. Use when the user invokes /planning:versioning or when an orchestrator calls this skill at the end of an implementation run. Requires an explicit bump level or --infer — empty input is an error.

Reviews completed work and updates or creates project documentation (README, CLAUDE.md, AGENTS.md, .claude/rules/, docs/) to prevent stale docs. Invoked at the end of an implementation workflow with a base ref; diffs the working tree against that base to capture every change on the branch — committed and uncommitted.

Rewrites technical text into plain English for a human reader. Use when an agent has produced prose that reads like implementation notes, or when the user asks to "humanize", "rewrite for humans", or "make this less technical" — for PR descriptions, ticket bodies, release notes, or code documentation. Returns only the rewritten text, ready to drop in.

Helps transform vague change requests into clear, structured requirements through targeted conversation. Use when a user has a change in mind for the codebase but hasn't fully articulated what they need — e.g. "we should add X", "improve Y", "fix the way Z works", "let's build a feature for...". Scales question depth to the complexity of the work. Produces an in-chat requirements summary suitable for handing to a ticket tool, a planner, another skill, or directly to an implementation agent.

Generates a detailed, actionable implementation plan for a non-trivial coding task. Invoked by the implementer (issue-implementation skill) at claim time when the claimed task does not meet the trivial-edit skip criteria. Returns an