全部能力

Web crawling and scraping tool with LLM-optimized output. 网页爬虫爬取工具 | Web crawler, web scraper, spider. DuckDuckGo search, site crawling, dynamic page scraping.…

Batch-generate images via OpenAI Images API. Random prompt sampler + `index.html` gallery.

firecrawl-deep-research — an installable skill for AI agents, published by firecrawl/firecrawl-workflows.

Search, install, and manage Agent Skills locally and from GitHub, then help decide whether the task really needs a skill or another customization primitive.…

autoresearch — 一个可安装的 AI 智能体技能，由 uditgoenka/autoresearch 发布。

为 Agentica/REPL 智能体编写避免 LLM 指令歧义的可靠提示词 核心原则 1. 明确执行模式 - 区分"思考模式"与"执行模式" - 使用显式标记如 `[THINK]` / `[EXECUTE]` 分隔推理与操作 2. 消除代词歧义 - 避免：它、这个、那个、上述 - 改用：精确的变量名、文件路径、函数签名 3. 结构化输出格式 - 强制指定响应格式（JSON/YAML/代码块） - 定义严格的字段顺序和必填项 4. 原子化操作指令 - 每条指令只包含单一动作 - 禁止复合句："读取并修改文件" → 拆分为两步 5. 状态显式传递 - 要求智能体在每次响应中重复关键上下文 - 避免依赖对话历史隐式记忆 具体模式 文件操作提示词模板 ``` 任务：修改文件 约束： - 目标路径：/absolute/path/to/file.ext - 操作类型：[READ | WRITE | PATCH | DELETE] - 若 PATCH，必须提供：原始片段（≥3 行上下文）、替换片段 输出格式： [CONFIRM] 路径：{path} 操作前哈希：{sha256} 操作后哈希：{sha256} 变更行数：{n} ``` REPL 代码执行提示词 ``` 环境：Node.js v20, 只读文件系统 /workspace 可用 API：{显式列出允许调用的模块} 执行规则： 1. 先输出 [PLAN] 步骤分解 2. 每步输出 [STEP n] 代码块，含超时设置 3. 禁止：eval、Function constructor、child_process 4. 副作用限制：仅允许写入 /tmp/output/* 错误处理： - 非预期异常 → 立即 [HALT] 并输出堆栈 - 预期异常 → [CATCH] 块显式处理 ``` 防注入模式 ``` 用户输入占位符：{{USER_INPUT}} 处理规则： - 所有 {{USER_INPUT}} 必须经 JSON.stringify 转义 - 禁止将用户输入拼接至模板字符串字面量 - 验证：输入长度 ≤ 4096，匹配预定义正则 ^[a-zA-Z0-9_]+$ ``` 状态机控制 ``` 当前状态：{IDLE | PENDING_CONFIRM | EXECUTING | ERROR} 允许转移： IDLE → PENDING_CONFIRM（需用户显式 [CONFIRM]） PENDING_CONFIRM → EXECUTING EXECUTING → IDLE | ERROR ERROR → IDLE（需 [RESET]） 禁止转移： 任何状态 → 绕过 PENDING_CONFIRM 的直接执行 ``` 验证清单 - [ ] 提示词是否包含"不要"类否定指令？改为正面约束 - [ ] 是否存在未量化的形容词（快速、大量、最近）？替换为具体数值 - [ ] 输出是否可被静态解析？避免自然语言包裹关键数据 - [ ] 边界情况是否覆盖：空输入、超长输入、特殊字符、并发请求 - [ ] 是否显式定义"完成"标准，而非依赖智能体推断 示例：安全的文件搜索任务 ``` [CONTEXT] 工作目录：/project/src .gitignore 已加载，排除路径：node_modules/, .git/, dist/ [TASK] 搜索所有引用 "deprecatedAPI" 的 TypeScript 文件 [CONSTRAINTS] - 文件扩展名限制：.ts, .tsx - 匹配模式：整词匹配，大小写敏感 - 最大结果数：50 - 单文件最大匹配行数：10 [OUTPUT_FORMAT] ```json { "search_id": "uuid-v4", "matches": [ { "file": "相对于工作目录的路径", "line": 整数, "column": 整数, "context": "前后各2行，共5行" } ], "truncated": true/false, "stats": {"files_searched": 整数, "files_matched": 整数} } ``` [VERIFICATION] 执行完成后，输出 [VERIFY] 标记并复述： - 实际扫描文件数 - 与 .gitignore 排除数 - 结果哈希：SHA256(matches 的 JSON 字符串) ```

firecrawl-lead-research — an installable skill for AI agents, published by firecrawl/firecrawl-workflows.

firecrawl-market-research — an installable skill for AI agents, published by firecrawl/firecrawl-workflows.

eastmoney_financial_search — an installable skill for AI agents, published by meission/eastmoney.

This skill provides comprehensive instructions for interacting with the Raindrop.io bookmarks service via its REST API using curl and jq. It covers authentication, CRUD operations for collections, raindrops (bookmarks), tags, highlights, filters, import/export, and backups. Use this skill whenever the user asks to work with their bookmarks from Raindrop.io, including reading, creating, updating, deleting, searching, or organising bookmarks and collections.

Invoke parallel document-specialist agents for external web searches and documentation lookup

Use when the user needs to run GitNexus CLI commands like analyze/index a repo, check status, clean the index, generate a wiki, or list indexed repos.…

Search for existing content pages containing a specific block in AEM Edge Delivery Services. Reports URLs with occurrences and variants to help identify test…

Configure AWS MCP servers for documentation search and API access. Use when setting up AWS MCP, configuring AWS documentation tools, troubleshooting MCP…

Analyze MSBuild binary logs to diagnose build failures by replaying binlogs to searchable text logs. Only activate in MSBuild/.NET build context. USE FOR:…

Generates comprehensive, workable unit tests for any programming language using a multi-agent pipeline. Use when asked to generate tests, write unit tests, improve test coverage, add test coverage, or create test files. Supports C#, TypeScript, JavaScript, Python, Go, Rust, Java, and more. Orchestrates research, planning, and implementation phases to produce tests that compile, pass, and follow project conventions. DO NOT USE FOR: running existing tests, executing dotnet test, applying test filters, detecting test platforms, or troubleshooting test execution (use run-tests for all of these).

Check which exchanges work from your location and search for tokens with trading rules (min order size, price increment, order types).

Guides systematic multi-pass review and editing of scientific manuscripts (research articles, reviews, perspectives) to improve clarity, structure, scientific…

llm-context — an installable skill for AI agents, published by brave/brave-search-skills.

Load first. State-today index of distilled Eve Horizon system docs with task-based routing for CLI/API usage, manifests, pipelines, jobs, secrets, agents,…