Starter Implement

Overview & Purpose

starter-implement is a starter workflow for for the primary agent loop: IMPLEMENT. It ensures disciplined execution of reasoning frameworks throughout this critical phase.

Execution Workflow

1. Import Verified Plan (session-import): Deserializes stored plan state from disk to initialize memory context.
2. Execute Step-by-Step (sequential-thinking): Trace execution steps linearly, allowing dynamic thought revisions when needed.
3. Code Dry-Run & State Inspection (code-execution-reasoning): Track memory, loop invariants, and pointer state line by line.
4. Metacognitive Self-Audit (metacognitive-monitoring): Monitor execution in real time to detect confidence drift or confirmation bias.
5. Systematic Debugging (debugging-approach): If an anomaly or test failure occurs, isolate root causes via binary search troubleshooting.
6. Serialize Execution State (session-export): Save final implementation state to disk.

Expected Output Contract

### Agentic Loop Phase: [STARTER-IMPLEMENT]
- **Phase Status**: [Complete / In Progress]
- **Key Deliverables**: [Summary of Phase Output]
- **Hand-off Artifact**: [Exported State / Implementation Code]

Scripts

• scripts/starter_implement.py - State serialization and execution helper for starter-implement.

Gotchas

• Ensure starter-plan has exported a valid plan state before triggering starter-implement.