Agent
Five specialists, plus the orchestrator. The orchestrator's prompt is the project root CLAUDE.md. Each specialist's prompt is at agents/ .md with a top-of-file CONTRACT block (HTML-commented YAML, parseable).
Agent
Agent "analyst_narrator" from ai-analyst-lab/agentxp, covering analyst-narrator, bundle, tools, output and discipline.
Agent
Agent "critic" from ai-analyst-lab/agentxp, covering critic, bundle, tools, judging modes and briefconsistency.
Agent
Agent "designer" from ai-analyst-lab/agentxp, covering designer, bundle, tools, output and discipline.
Agent
Agent "sql_specialist" from ai-analyst-lab/agentxp, covering sql specialist, bundle, tools, output and discipline.
Agent
Agent "understander" from ai-analyst-lab/agentxp, covering understander, bundle, tools, output and voice.
ai-analyst-lab/ai-analyst-starter
Instructions file
Claude Code instructions for ai-analyst-lab/ai-analyst-starter, covering claude.md -- ai analyst (starter), who you are, your skills, your agents and available data.
ai-analyst-lab/ai-analyst-starter
Agent
Every agent .md file MUST begin with a CONTRACT block -- a YAML declaration inside an HTML comment that describes the agent's interface. The OR-3 DAG walker reads these contracts to build the execution graph.
ai-analyst-lab/ai-analyst-starter
Agent
Agent "data-explorer" from ai-analyst-lab/ai-analyst-starter, covering agent: data explorer, purpose, inputs, workflow and step 1 — list what is there.