writing-plans（写作计划）

Writing Plans

Overview

Write comprehensive implementation plans assuming the engineer has zero context for our codebase and questionable taste. Document everything they need to know: which files to touch for each task, code, testing, docs they might need to check, how to test it. Give them the whole plan as bite-sized tasks. DRY. YAGNI. TDD. Frequent commits.

Assume they are a skilled developer, but know almost nothing about our toolset or problem domain. Assume they don't know good test design very well.

Announce at start: "I'm using the writing-plans skill to create the implementation plan."

Context: This should be run in a dedicated worktree (created by brainstorming skill).

Save plans to: docs/plans/YYYY-MM-DD-<feature-name>.md

Experience-Informed Planning (MANDATORY)

Before finalizing any plan, query past experience from neo4j-agent-memory to avoid repeating mistakes and leverage proven patterns.

Step 1: Extract concepts. Identify 3-5 key technical concepts from the plan (technologies, patterns, architectural choices).

Step 2: Query the knowledge graph. For each concept, run:

memory_search(query="[concept]", memory_types=["entities"])

Then run a scoped Cypher query for cross-project principles:

graph_query: MATCH (p:Entity)
  WHERE p.type = 'OBJECT' AND p.subtype IN ['PATTERN', 'GOTCHA']
    AND p.metadata IS NOT NULL
  WITH p, apoc.convert.fromJsonMap(p.metadata) AS meta
  WHERE any(tag IN meta.stack_tags WHERE tag IN $plan_stack_tags)
    AND meta.confidence IN ['validated', 'derived']
    AND (meta.scope = 'cross-project' OR meta.scope = $current_project_scope)
  RETURN p.name, p.description, meta.confidence, meta.source_lineage
  ORDER BY CASE meta.confidence WHEN 'validated' THEN 0 ELSE 1 END
  LIMIT 10

Replace $plan_stack_tags with the plan's tech stack (e.g., ['python', 'postgresql', 'fastapi']) and $current_project_scope with 'project:[name]'.

Step 3: Incorporate. If any results contradict or inform the plan, add constraints or warnings to the relevant tasks. Note which principles were applied in the plan header under **Experience Applied:**.

Skip if neo4j-agent-memory is unavailable. Do not block plan generation on graph queries.

Bite-Sized Task Granularity

Each step is one action (2-5 minutes):

• "Write the failing test" - step
• "Run it to make sure it fails" - step
• "Implement the minimal code to make the test pass" - step
• "Run the tests and make sure they pass" - step
• "Commit" - step

Plan Document Header

Every plan MUST start with this header:

# [Feature Name] Implementation Plan

> **For Claude:** REQUIRED SUB-SKILL: Use software-forge:executing-plans to implement this plan task-by-task.

**Goal:** [One sentence describing what this builds]

**Architecture:** [2-3 sentences about approach]

**Tech Stack:** [Key technologies/libraries]

---

Task Structure

### Task N: [Component Name]

**Files:**
- Create: `exact/path/to/file.py`
- Modify: `exact/path/to/existing.py:123-145`
- Test: `tests/exact/path/to/test.py`

**Step 1: Write the failing test**

def test_specific_behavior(): result = function(input) assert result == expected

**Step 2: Run test to verify it fails**

Run: `pytest tests/path/test.py::test_name -v`
Expected: FAIL with "function not defined"

**Step 3: Write minimal implementation**

def function(input): return expected

**Step 4: Run test to verify it passes**

Run: `pytest tests/path/test.py::test_name -v`
Expected: PASS

**Step 5: Commit**

git add tests/path/test.py src/path/file.py git commit -m "feat: add specific feature"

Remember

• Exact file paths always
• Complete code in plan (not "add validation")
• Exact commands with expected output
• Reference relevant skills with @ syntax
• DRY, YAGNI, TDD, frequent commits

Output Protocol (MANDATORY)

After generating the full plan:

1. Write the complete plan to docs/plans/YYYY-MM-DD-<topic>-plan.md
2. Display ONLY a summary table to the conversation:

1. End with: "Full plan saved to docs/plans/[filename]. [N] tasks, [M] files, [T] tests."

Do NOT output the full plan to the conversation. The plan lives on disk.

Execution Handoff

After saving the plan, offer execution choice:

"Plan complete and saved to docs/plans/<filename>.md. Two execution options:

1. Subagent-Driven (this session) - I dispatch fresh subagent per task, review between tasks, fast iteration

2. Parallel Session (separate) - Open new session with executing-plans, batch execution with checkpoints

Which approach?"

If Subagent-Driven chosen:

• REQUIRED SUB-SKILL: Use software-forge:subagent-driven-development
• Stay in this session
• Fresh subagent per task + code review

If Parallel Session chosen:

• Guide them to open new session in worktree
• REQUIRED SUB-SKILL: New session uses software-forge:executing-plans