Implement Plan

Execute a structured implementation plan with normalize → test → commit cycle

DYAI2025 updated 2mo ago
Claude CodeGeneric
View source ↗
---
description: Execute a structured implementation plan with normalize → test → commit cycle
allowed-tools: Read, Write, Edit, Bash, Glob, Grep, Task, TaskCreate, TaskUpdate, TaskList
---

## Context
LeanDeep-annotator follows a strict plan → implement → normalize → test → commit workflow.
Source-of-truth marker YAMLs live in `build/markers_rated/` (NOT `build/markers_normalized/`).
The normalizer overwrites `build/markers_normalized/` on every run.

## Your Task

Execute the implementation plan provided by the user. The plan is either:
- Pasted directly in the message
- Referenced as a file path (read it first)
- The most recent plan from docs/plans/

### Steps

1. **Read the plan** — understand all files to create/modify, the order, and success criteria
2. **Read all existing files** that will be modified (never edit blind)
3. **Create TaskList** — one task per file/component in the plan
4. **Implement** — file by file, following the plan's order
5. **Normalize** (if any marker YAML was changed):
   ```bash
   python3 tools/normalize_schema.py
  1. Run tests:
    python3 -m pytest tests/ -x -q
    
  2. Fix any failures — iterate until green
  3. Report — summary of what was implemented, test results

Guardrails

  • NEVER edit files in build/markers_normalized/ directly — edit in build/markers_rated/ and run normalizer
  • ALWAYS read a file before editing it
  • ALWAYS run the full test suite, not just new tests
  • If the plan includes new endpoints, verify backward compatibility (existing endpoints unchanged)
  • Do NOT commit or push unless the user explicitly asks

Generated by /reflect-skills from 5 session patterns ```

Maintain Implement Plan?

Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.

[Implement Plan on getagentictools](https://getagentictools.com/loops/dyai2025-implement-plan?ref=badge)
npx agentictools info loops/dyai2025-implement-plan

The second line is the CLI lookup for this page — handy in READMEs and docs.