.agent.md — Investor Flow project

Autopilot

Autopilot: Enabled

Routing

IF task type = research → load prompts/research.md + .rules.md IF task type = design → load prompts/design.md + SPEC.md IF task type = test_design → load prompts/test_design.md + SPEC.md + DESIGN.md IF task type = implement → load prompts/implement.md + SPEC.md + DESIGN.md + TEST_PLAN.md + CONTRACT.md IF task type = code_review → load prompts/code_review.md + SPEC.md + DESIGN.md + IMPLEMENTATION.md IF task type = bug_find → load prompts/bug_finder.md + SPEC.md + code IF task type = doc_review → load prompts/doc_review.md + DESIGN.md IF task type = orchestrate → load prompts/orchestrate.md + project structure

State Enforcement (v2.0)

All phase transitions must go through status.py (always pass --project):

Create: status.py --create-task {name} --project {project}
Transition: status.py --transition {phase} --task {name} --project {project}
Approve: status.py --approve --task {name} --project {project}
Audit: status.py --audit --project {project}

Agent Configuration

Mode: multi-agent Agents:

id: ornith-35 phases: [research, decomposition, design, test_design, implement, code_review] role: implementer
id: qwopus35b phases: [research, decomposition, design, test_design, implement, code_review] role: implementer
id: orchestrator phases: [new, complete, human_intervention] role: coordinator Lock timeout: 30m

Team (models — .automaton/models.json)

ornith-35 (omlx/local, 131K) — implementer (primary; fast)
qwopus35b (remote, 131K) — implementer (cross-reviews ornith-35's code; ornith-35 cross-reviews qwopus35b's)
orchestrator — coordination ONLY (no code writing, no code review); dispatches implementers, assigns cross-reviews, enforces conflict-of-interest (no implementer reviews its own code).

1.9 KiB Raw Blame History