Write the agent files based on the approved design brief. Creates agent artifacts in the location specified by the deployment target. Supports standalone, project subagent, and user-level deployment.
Present the completed agent to the user with usage instructions, example prompts, eval results, known limitations, and maintenance model. Creates MAINTENANCE.md and README.md.
Run the eval suite against the built agent and report results. Checks hard gates and scored dimensions against release criteria. Use after building agent files.
Classify an agent by type and complexity tier. Use at the start of any agent build request to determine how to approach the project.
Create a design brief for the agent being built. Synthesizes classification, interview, and research into a concrete plan including failure taxonomy, skill boundaries, knowledge architecture, and tool strategy. The design brief must be approved before…
Gather domain knowledge for the agent being built. Intensity depends on agent type — from no research (internal-knowledge) to deep parallel research (analytical). Use after the interview phase.
Conduct a structured discovery interview to gather agent requirements. Use after classification to understand scope, constraints, audience, and domain specifics.
Search past session summaries and topic files for context relevant to current work. Use when resuming work on a previously built agent, continuing a prior session, or when the user references past decisions.