prompt-engineering
3
总安装量
3
周安装量
#57889
全站排名
安装命令
npx skills add https://github.com/d-kimuson/dotfiles --skill prompt-engineering
Agent 安装分布
opencode
3
claude-code
3
github-copilot
3
codex
3
kimi-cli
3
gemini-cli
3
Skill 文档
Guidelines for creating and editing Coding Agent prompts (commands, agents, skills, context files).
Important: Always Read the detailed reference for the corresponding prompt type before starting work.
Prompt Type Quick Reference
| Type | Location | Invocation | Purpose | Reference |
|---|---|---|---|---|
| Command | .claude/commands/<name>.md |
/command-name |
Reusable tasks for users | references/command.md |
| Agent | .{.super-agent,.claude}/agents/<name>.md |
@agent-name / Task tool |
Specialized subagents | references/agent.md |
| Skill | `{.super-agent | .claude | .codex | .github}/skills//SKILL.md` |
| Context File | .claude/CLAUDE.md etc. |
Auto-loaded | Always-needed project context | references/context-file.md |
| Document | Any | Manual reference | Standalone prompts | – |
Other references:
references/orchestration.md– Orchestrator design for calling subagentsreferences/permission-syntax.md– Permission syntax for allowed-toolsreferences/hooks.md– Lifecycle hook configuration
Core Principles
1. Single Responsibility
Each prompt has one clear purpose.
- â Environment setup only / Code implementation only
2. Independence from Caller
Avoid references to “orchestrator” or “parent task”; focus on input/output contracts.
- â “Analyze the provided code and identify issues…”
3. Conciseness
Only information necessary for execution. Remove verbose examples, hypothetical paths, and generic patterns.
4. Noise Avoidance
- Avoid examples in multiple languages (choose the primary language)
- Avoid hypothetical file paths (CONTRIBUTING.md, etc.)
- Omit detailed steps that the LLM can infer
Formatting Rules
- No h1 headings: Do not start with
# - Language:
descriptionin the project’s primary language; body can also be in primary language - XML tags: Use for structuring when there are multiple sections
Orchestration
When calling subagents:
- Invocation template required: Include complete Task tool usage example
- Responsibility separation: Subagents should be generic; task-specific details go in templates
Special Rules for Context Files
Be especially careful since they are always loaded:
- 80% rule: Only information needed by 80% of tasks
- Index-first: Reference details via pointers
- Scrutinize commands: Only commands the LLM will autonomously execute
- Target under 200 lines