These are commonly used system-provided agents. This is not an exhaustive list — to see all

Slack bot using `@slack/bolt` with Socket Mode. Entry point: `src/index.ts`

../../.agents/claude/commands/setup-agents.md

Seek out and update agent discovery files as conventions evolve. Use when agent ecosystems add new ways to state behavior, skills, or rules; when discovery files need revision; or when maintaining FOR_AGENTS.md, .cursorrules, docs/for-agents.md, docs/agent-discovery.md.

/Users/hansel/conductor/repos/forbotsake/.claude/skills/gstack/open-gstack-browser/SKILL.md

他山 Cursor Skill 体系（合集入口）— 95 个 Skills、32 条 Rules、18 个 SubAgents；覆盖产品→架构→开发→测试→上线全链路，含认知积累与 Skill 自进化机制。适用于 Cursor / OpenClaw 工作区一键安装。触发词：他山 skill、Tashan skills、tashan-cursor-skills、安装他山规范、三大闭环、session-bootstrap。

捕捉碎片化思考，结构化写入认知结构L2层。触发词：「碎片」「记录一下」「想法」「新洞见」「我发现」「我觉得」「随手记」「想记一个」。

任意输入的路由分类器。在输入意图不明确时，判断应走「认知更新路径A（cognitive-capture-fragment）」还是「任务执行路径B（对应任务Skill）」，输出明确路由结论 + 判断理由，不自动触发下游。触发词：「帮我判断这是认知更新还是任务执行」「这个输入是什么类型」「先分类再执行」「路由分类一下」。不应触发：关于计算机网络路由算法的技术问题。

从整个工作区出发，系统性地重组认知结构并整理全局文档归属。适用于首次建立认知结构、或积累了大量散落文档需要批量归档时。基于三大闭环框架（Loop1 Skill体系 / Loop2 思维体系 / Loop3 场景投射）做全工作区分类路由。触发词：「系统整理」「重组认知结构」「从文档构建认知结构」「整合文档」「全量梳理」「认知结构乱了」「全工作区整理」「文档归属梳理」「如何加入认知结构」「怎么加入认知结构」「按规范加入」「未更新的文档怎么处理」「这些文档要怎么整理」「有没有关于X的已有文档」。注意：若用户仅询问流程而非执行重组，执行 Step 0.5（检查分类清单现状）后，先向用户展示已有文档状态和分类建议，再确认是否执行完整重组。

指导 AI 正确修改①公理体系或②体系结构（A-C-B三粒子）。当用户提到「修改公理」「修改体系结构」「发现规则漏洞」「增加粒子」「修改规则」时触发。对应 AGENT_RULES.md RULE-13~16 和 08_技术体系修改规范.md。

对任何主题进行系统调研，自动产出图文并茂的 Markdown 文档（含 Mermaid 结构图 + qwen-image 可视化），保存到认知结构对应维度的知识库，并自动注册到文档分类清单和知识图谱。当用户说「帮我调研/调研一下/系统调研/研究一下/帮我系统了解/深度研究 [主题]」时触发。

AI工程师角色。关键词：AI/LLM/智能体/Prompt/上下文工程/多智能体/工作流/RAG/Function Call/效果评测。激活后设计提示词、构建智能体调用链、评测输出质量。

产品经理角色。关键词：产品/需求/用户流/闭环/MVP/产品定义/开发计划/用户动线/功能设计/产品迭代。激活后立即读产品定义.md和开发计划.md，执行沙盘推演关卡A。

Skill/Agent/Rule 开发全流程双视角导引者。新建或重构任何 Skill/Agent/Rule 时激活。判断复杂度级别（L1-L4），引导完成双层产品定义（人层意图 + Agent层规格），按级别触发对应关卡（A=AI执行者模拟/B=系统冲突破坏/C=行为验证）。触发词：「新建Skill」「设计一个Skill」「要写一个新的Skill」「新建Agent」「设计子智能体」「新建Rule」「设计规则」。

全量搜索外部 Skill/Agent/Rule 仓库，必须执行7维度协议，禁止以少于7维度的搜索声称「全量完成」。触发词：「全量搜索 skill 库」「搜索 GitHub skill 仓库」「找所有 skill 库」「有没有遗漏的 skill 仓库」「更新 skill 参考库」。

修改 Skill / Rule / Agent 文件的操作规范。当 AI 或用户需要新增、修改、删除 .cursor/skills/ 或 .cursor/rules/ 或 .cursor/agents/ 中的任何文件时触发。修改前必须遵循本规范，未经本规范允许不可直接改写 Skill/Rule/Agent。

指导 AI 正确填写任务日志，并在任务完成后自动提取认知价值。当任务完成后、对话即将结束时、或用户提到「写日志」「更新日志」「任务记录」时自动触发。对应 AGENT_RULES.md RULE-01/02/03。版本：v2.0（2026-03-19新增步骤四：认知提取）

| | 产品反馈处理（本 Skill）| 规范审核迭代 |

Everything OpenCode (EOC) is a comprehensive OpenCode plugin providing agents, commands, skills, hooks, and rules for AI-assisted software development.

This file contains conventions and instructions for AI agents (Claude, etc.)

Before committing code, ask the user if you should bump the version of the plugin by one patch number. The version is stored at plugins/brad-personal/.claude-plugin/plugin.json.

Use loom workspace to create/manage worktrees for isolated execution of tickets.

循环执行的 Skill。当需要重复执行直到满足条件时触发。触发词：循环、loop、重复、直到、until、持续。

You are an expert in multi-period lot-sizing models and production/inventory planning optimization. Your goal is to help determine optimal order or production quantities across multiple time periods t

{{SKILL_DESCRIPTION}} Use this skill when working with core functionality for LÖVE games.

Imported skill __main__ from langchain

LaTeX 模板高保真优化器，支持任意 LaTeX 模板的样式参数对齐、标题文字对齐、标题格式对比（加粗）、HTML 可视化报告、LaTeX 自动修复建议和像素级 PDF 对比验证

maker-orchestrator

Run configuration handling for persistent command configuration storage

Standardizes date handling and timezone logic (UTC/PKT) for travel bookings. Use when displaying or saving tour dates.

Manages parallel conversation branches using git worktrees. Use when you want to explore multiple solution approaches simultaneously, need to try different implementations, or want to checkpoint your current conversation state before branching into alternative paths.

|-

Analyze codebase with parallel mapper agents

Scans and reports complete marketplace inventory (bundles, agents, commands, skills, scripts)

Automates updates to the Functional Area Master Index when new features are added or documentation changes. Maintains the single navigation source for all feature documentation, ensuring discoverability and organization.

当用户提供飞书 MCP URL 并请求配置时触发。自动同时配置到 Claude Code 和 Cursor，简化团队成员的环境设置流程。

Provides score interpretation (< 0.3 good, > 0.5 skip), query optimization, and source naming for query_documents, ingest_file, ingest_data tools. Use this skill when working with RAG, searching documents, ingesting files, saving web content, or handling PDF, HTML, DOCX, TXT, Markdown.

MCP 服务器智能管理助手。自动检测 MCP 可用性、智能开关、功能问答，提供人性化的 MCP 管理体验。

Create, modify, and manage MCP tools for the Naki mahjong AI assistant. Use when adding new MCP tools, modifying existing tools, or fixing tool-related issues. This skill understands the Protocol-based MCPTool architecture.

Match SAE features between Ultra (24K) and Full (2K) models based on activation patterns and token overlap

All things needed for interacting with medical records

Resolve common merge conflicts in merge-queue worktree including observations.jsonl, uv.lock, pyproject.toml, and egg-info files

Two distinct features for managing message context:

methodology

methodology-system

mgrep is a semantic search tool that enables natural language queries across code, text, PDFs, and images. It is particularly effective for exploring larger or complex codebases where traditional patt

Iterative development loop that feeds the same prompt back after each iteration until task completion. Use /bluera-base:milhouse-loop to start.

Launch ralph.sh in background. Best for large features (10+ tasks). Automatic execution until completion.

Select feature, select mode, delegate to execution skill.

> **创建日期**: 2026-01-15