deepcode-self-refer
Answers questions about Deep Code CLI itself — including features, configuration options, slash commands, Skills, MCP integration, permissions, notifications, session persistence, and troubleshooting. Use this when users ask how to configure or use Deep Code, how to set up an MCP server, configure notifications (such as Slack/Feishu), manage permissions, view available skills, understand slash commands, configure thinking mode, etc.
lessweb · 已有 AI 审核记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add lessweb/deepcode-cli --skill deepcode-self-refer技能说明
name: deepcode-self-refer description: Answers questions about Deep Code CLI itself — including features, configuration options, slash commands, Skills, MCP integration, permissions, notifications, session persistence, and troubleshooting. Use this when users ask how to configure or use Deep Code, how to set up an MCP server, configure notifications (such as Slack/Feishu), manage permissions, view available skills, understand slash commands, configure thinking mode, etc.
Deep Code Self-Refer
This Skill helps you answer user questions about Deep Code CLI itself by consulting the reference documentation bundled with this Skill. All docs live in the references/ subdirectory — always refer to them for authoritative answers.
When to use this Skill
Use this Skill when the user asks any question about Deep Code itself, such as:
- "列出可用的 skills"
- "如何配置 MCP?"
- "给当前项目配置 playwright mcp"
- "怎么启用搜索功能?"
- "支持哪些模型?"
- "如何配置思考模式?"
- "怎么设置权限?"
- "任务完成后怎么发通知?"
- "支持哪些斜杠命令?"
- "会话历史保存在哪里?"
- "/undo 是怎么工作的?"
- "Deep Code 和 VSCode 插件怎么配合?"
- Any other question about Deep Code CLI's features, configuration, or usage.
Instructions
Step 1: Identify the topic
Map the user's question to the appropriate document(s):
| Topic | Document | Key contents |
|---|---|---|
| Overview, features, quick start | references/README.md | Installation, slash commands, keyboard shortcuts, supported models, FAQ |
| Configuration & settings | references/configuration.md | settings.json fields, config hierarchy, env vars, thinking mode, reasoning effort, webSearchTool, enabledSkills |
| MCP setup & usage | references/mcp.md | MCP server config format, GitHub/Playwright/Filesystem examples, tool naming (mcp__<name>__<tool>), troubleshooting |
| Permissions | references/permission.md | Permission scopes (12 types), allow/deny/ask/defaultMode config, priority rules, persistence |
| Notifications | references/notify.md | Notify script path, injected env vars, Slack/Feishu/iTerm2/macOS/Linux/Windows examples |
| Session persistence | references/session-persistence.md | Storage paths, JSONL format, session index, compaction, /undo mechanics, code snapshots |
Step 2: Read the relevant document(s)
Use the Read tool to read the appropriate document(s) from the list above. All paths are relative to this Skill's loaded root directory, where the references/ subdirectory lives.
- If the question spans multiple topics, read multiple documents.
- If a document doesn't exist in the user's preferred language (e.g., Chinese), try the other language variant (e.g.,
references/configuration_en.md). - When answering from references/README.md, focus on the relevant sections.
Step 3: Answer with precision
- Quote the doc directly for config examples, JSON snippets, or command syntax.
- Don't guess — if the answer isn't in the docs, say so and suggest checking GitHub Issues.
- Provide copy-paste-ready configurations when the user asks to set something up (e.g., MCP servers, notify scripts, permissions).
- Mention related docs when appropriate (e.g., MCP setup references
references/mcp.md, the permissions section referencesreferences/permission.md).
Step 4: Handle common request patterns
"列出/查看可用的 skills":
- Treat
/skillsas the canonical UI for listing currently available skills. - If answering directly, do not infer the list only from loaded skill prompts or from project/user directories. Enumerate all discovery roots:
./.deepcode/skills/<folder>/SKILL.md./.agents/skills/<folder>/SKILL.md~/.deepcode/skills/<folder>/SKILL.md~/.agents/skills/<folder>/SKILL.md- bundled built-in skills as
bundled:<folder>/SKILL.md
- For a source checkout, bundled skills live under
templates/skills/bundled/<folder>/SKILL.md. For a packaged install, bundled skills may live underdist/bundled/<folder>/SKILL.md. - Read each candidate
SKILL.mdfrontmatter to get the resolvednameanddescription; the folder name is only a fallback. - De-duplicate by resolved
name, keeping the highest-priority root from the order above. - Apply
enabledSkillsfromsettings.json: ifenabledSkills["<name>"] === false, do not list that skill as available. - Clearly separate discoverable skills from other concepts:
- Discoverable skills are selectable through
/skillsand come from the roots above. - Bundled skills are discoverable skills shipped with Deep Code, such as
bundled:deepcode-self-refer/SKILL.md. - Default prompt templates or always-injected guidance are not necessarily discoverable skills unless they also exist as
*/SKILL.mdin one of the scan roots. - Slash commands such as
/skills,/mcp, and/undoare commands, not skills.
- Discoverable skills are selectable through
- Mention that
/skillscan be used to verify the result andenabledSkillscan enable/disable specific skills by name.
"配置 MCP":
- Read
references/mcp.mdfor the MCP format and examples - Ask the user for any required credentials (e.g., GitHub token)
- Provide the exact
mcpServersJSON block to add tosettings.json - Mention using
/mcpto verify the setup afterwards
"如何配置/修改 <设置项>":
- Read
references/configuration.md - Explain which
settings.jsonfield controls the setting - Clarify user-level (
~/.deepcode/settings.json) vs project-level (`.deepco