prp

Comprehensive PR code review - checks diff, patterns, runs validation, comments on PR. Single-pass by default; pass --agents for a multi-agent specialist fan-out (comments, tests, errors, types, code, docs, simplify). Use when the user wants to review a pull request, wants a thorough/multi-agent review, or invokes $prp-review.

Interactive PRD generator - problem-first, hypothesis-driven product spec. Use when the user wants to create a PRD, write a product or feature spec, scope a new product, or invokes $prp-prd.

Create a PR from current branch with unpushed commits. Use when the user wants to open or create a pull request, push and PR a branch, or invokes $prp-pr.

Create comprehensive feature implementation plan with codebase analysis and research. Also wires bidirectional back/forward references between existing plans (the update-references workflow). Use when the user wants to plan a feature, turn a PRD into an implementation plan, "link two plans", "add a back/forward reference", "connect related plans", or invokes $prp-plan.

Turn the current session into an SDLC orchestrator that coordinates parallel background agents running PRP skills in isolated worktrees - decompose work into workstreams, launch and steer background agents with your harness's delegation tools, hold review gates as the user's proxy, and sequence merges. Use when the user wants to "spawn N agents in separate worktrees", "run prp-issue on these issues in parallel", "orchestrate these features", "act as my orchestrator", "coordinate agents through the PRP pipeline", "ship these issues in parallel", or invokes $prp-orchestrate.

Author new Claude Code skills PRP-style, and refactor existing fat skills into a lean SKILL.md plus references/. Use when the user wants to "create a skill", "write a new skill", "turn a command into a skill", "split a skill into references", "trim a SKILL.md", "extract output templates from a skill", "make a skill leaner", or invokes $prp-meta-skill.

Run the autonomous cyclic PRP pipeline end to end (plan → implement → pr → review, looping review→fix until the PR is clean). Use when the user wants to "ship feature X end to end", "run the full PRP loop", "auto-implement and open a PR for a feature", or invokes $prp-loop.

Investigate a GitHub issue and implement the fix - analyze codebase, create a plan, then code, PR, and self-review. Use when the user wants to investigate or triage a GitHub issue or bug report, fix an investigated issue, implement an issue fix, or invokes $prp-issue.

Execute an implementation plan with rigorous validation loops. Use when the user wants to implement or execute a plan file, build a planned feature, or invokes $prp-implement.

Generate a mermaid-only visual supplement for a plan artifact - the data-model, architecture, and flow diagrams a human wants when absorbing a plan. Use when the user wants to "diagram this plan", "visualize the plan", "draw the plan's architecture", "make diagrams for this plan", or invokes $prp-diagram.

Deep root cause analysis - finds the actual cause, not just symptoms. Use when the user reports a bug, error, or stacktrace, wants to debug or do root-cause analysis, or invokes $prp-debug.

Quick commit with natural language file targeting. Use when the user wants to commit changes, stage and commit specific files, or invokes $prp-commit.

Research codebase questions using parallel agents - documents what exists, not what should change. Use when the user asks how the codebase works, where something lives, or invokes $prp-codebase-question.