Faion Writing Implementation Plan
Productivityfaion-writing-implementation-planFaion Writing Implementation Plan is an Agent Skill for productivity published in the majiayu000/claude-skill-registry repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
SDD Framework: Creates implementation-plan.md from design.md. Decomposes into tasks that fit 100k token context. Triggers on "implementation plan", "impl plan".
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Inspect the Agent Skill at https://github.com/majiayu000/claude-skill-registry/blob/HEAD/skills/development/faion-writing-implementation-plan/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/faion-writing-implementation-plan, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
SDD: Writing Implementation Plans
Communication: User's language. Docs: English.
Key Principle
Each task MUST fit 100k token context = atomic, single-agent executable.
Workflow
LOAD CONTEXT → ANALYZE COMPLEXITY → DEFINE WORK UNITS → APPLY 100k RULE → DEPENDENCIES → DRAFT → REVIEW → SAVE
Phase 1: Load Context
Read: constitution.md, spec.md, design.md
Phase 2: Analyze Complexity
For each file in design.md:
| File | Action | Complexity | Est. Tokens |
|---|---|---|---|
| models.py | CREATE | High | 40k |
| services.py | CREATE | High | 50k |
Factors: LOC, dependencies, business logic, tests
Phase 3: Define Work Units
Group related work:
## Work Unit 1: Data Layer
- models.py (CREATE)
Est: 40k tokens
Phase 4: Apply 100k Rule
Research: ~20k (read existing code)
Task file: ~5k
Implementation: ~50k
Buffer: ~25k
TOTAL < 100k
If > 100k: Split into smaller tasks.
Phase 5: Dependencies
| Task | Depends On | Enables |
|---|---|---|
| TASK_001 | — | TASK_003 |
| TASK_002 | — | TASK_003 |
| TASK_003 | 001, 002 | — |
Rules: No cycles, maximize parallelization.
Phase 6: Draft
Show section by section:
- Overview (total tasks, critical path, est. tokens)
- Tasks list (files, deps, FR/AD coverage, AC)
- Execution order (phases, parallel/sequential)
Phase 7: Review
Call faion-impl-plan-reviewer-agent agent. Checks: 100k compliance, deps, coverage.
Phase 8: Save
aidocs/sdd/{project}/features/{feature}/implementation-plan.md
Token Estimation Guide
| Component | Tokens |
|---|---|
| Django model simple | 5-10k |
| Django model complex | 15-25k |
| Service class | 20-40k |
| ViewSet | 15-30k |
| Test file | 20-40k |
Rule: If uncertain, estimate higher and split.
Output
implementation-plan.md → Next: faion-make-tasks skill (via /faion-net → create tasks)