基于 SOC 职业分类
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/roojs/OLLMchat --skill writing-plans命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
This skill should be used when the user needs to solve a complex task and wants a detailed execution plan using the best available resources. Analyzes user requirements, discovers available plugins/agents/skills/MCPs, performs intelligent matching with confidence scoring, and creates strategic execution plans with alternatives. Works across all AI CLI platforms.
This skill should be used when the user needs pre-implementation ideation for creative work, such as creating features, building components, adding functionality, or modifying behavior.
This skill should be used when the user needs to debug broken behaviour or review changed code for real defects. Follow the project bug-fix process (reproduce, bug log, root cause, propose with fences, approval before apply). Borrow a compact review checklist for diffs.
| name | writing-plans |
| description | This skill should be used when the user has a spec or requirements for a multi-step task before touching code. |
| license | MIT |
Create executable, low-ambiguity implementation plans that another engineer can run task-by-task with predictable outcomes.
Use this skill when:
Display progress at each planning phase:
[████░░░░░░░░░░░░░░░░] 25% — Phase 1/4: Gathering Context & Constraints
[████████░░░░░░░░░░░░] 50% — Phase 2/4: Decomposing Into Tasks
[████████████░░░░░░░░] 75% — Phase 3/4: Specifying Files & Commands
[████████████████████] 100% — Phase 4/4: Writing & Saving Plan
executing-plansWrite 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."
Tools: read / write / edit / bash (and codebase_search when researching the tree). Ask clarifying questions in chat.
Save plans to: docs/plans/YYYY-MM-DD-<feature-name>.md (or the path the user names).
Each step is one action (2-5 minutes):
Every plan MUST start with this header:
# [Feature Name] Implementation Plan
> **For the executor:** Use the `executing-plans` skill 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]
---
Every plan MUST include a ## To-do section listing the tasks the executor will walk. Use checkboxes (or ⏳ / ✔️ / ✅ if matching project plan style). executing-plans updates this section as work proceeds.
Example:
## To-do
- [ ] Task 1: …
- [ ] Task 2: …
- [ ] Task 3: …
Keep To-do items aligned with the detailed ### Task N sections below.
### Task N: [Component Name]
**Files:**
- Create: `exact/path/to/file.vala`
- Modify: `exact/path/to/existing.vala`
- Test: `tests/exact/path/to/test`
**Step 1: Write the failing test**
(concrete code or commands)
**Step 2: Run test to verify it fails**
Run: `…`
Expected: FAIL with …
**Step 3: Write minimal implementation**
(concrete code)
**Step 4: Run test to verify it passes**
Run: `…`
Expected: PASS
**Step 5: Commit** (only if the user wants commits)
```bash
git add …
git commit -m "…"
```
## To-doAfter saving the plan, offer:
"Plan complete and saved to docs/plans/<filename>.md. Continue in this session with executing-plans, or open a new session pointed at that file?"
## To-do.