Collaboratively discuss designs, evaluate trade-offs, and explore alternative implementation approaches. Starts from the simplest viable option and explores justified alternatives without fixed personas.
Implements an approved code-change plan safely. Use when beginning a feature, bug fix, or refactor that requires source changes, tests, a dedicated branch, small logical commits, and a GitHub pull request when applicable.
Write a right-sized, no-placeholder implementation plan as a series of focused tasks, with exact file paths, implementation details, proportionate test steps, and commit instructions. Performs a pre-planning preflight investigation and Ponytail decision…
Log items to come back to later — bugs found mid-task, feature ideas, project feedback — as GitHub Issues. Use when the user asks to log, list, view, or work through items saved for later.
Walk the developer through an implementation plan step-by-step in the chat interface. Explains the rationale, non-trivial alternatives considered, and risks for each step before pausing for feedback or approval.
Review the pull request snapshot prepared by agent-runner. Use only inside the agent-runner Docker image; it gathers GitHub context and applies the canonical code-review standard to the runner's pinned, isolated checkout.
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".
Explain the core concepts, library design patterns, system architecture, or host environment internals involved in a plan or git diff.