← 返回 Skills 市场
1134
总下载
0
收藏
1
当前安装
3
版本数
在 OpenClaw 中安装
/install unifuncs-deep-search
功能描述
Use UniFuncs Deep Search API for fast, comprehensive information gathering. Use this skill when users ask for deep search, broad investigation, or in-depth t...
安全使用建议
This skill appears to be a legitimate client for UniFuncs Deep Search, but the package metadata fails to declare the required UNIFUNCS_API_KEY — the SKILL.md and the bundled scripts do expect that API key. Before installing: (1) avoid sending passwords, secrets, or personally-identifiable data as query text (the queries and results are sent to a third-party API and may be published if options like push-to-share / set-public are used); (2) verify the publisher/source (there's no homepage or repository linked); (3) confirm you are comfortable storing/using an external API key and that the key has limited scope and revocation ability; (4) ask the publisher to update the manifest to declare UNIFUNCS_API_KEY as a required credential; and (5) if in doubt, inspect the included Python files (they are bundled) and test with non-sensitive queries first.
功能分析
Type: OpenClaw Skill
Name: unifuncs-deep-search
Version: 0.0.5
The skill bundle provides a legitimate interface for the UniFuncs Deep Search API, including scripts for task creation, status querying, and streaming reports (deep-search-report.py). The code is well-structured and uses standard Python libraries (urllib) to communicate exclusively with api.unifuncs.com. It implements a background worker pattern using subprocess.Popen to handle long-running search operations that exceed typical agent timeouts, which is a documented functional requirement. No evidence of data exfiltration, malicious persistence, or prompt injection was found; the background process is transparently managed and includes instructions for the user to resume reading results.
能力评估
Purpose & Capability
The name/description, SKILL.md, and included Python scripts are coherent: they implement a UniFuncs Deep Search client that calls https://api.unifuncs.com/deepsearch/* endpoints. That capability matches the stated purpose. However, the skill package metadata declares no required environment variables or primary credential while both SKILL.md and the code require an API key (UNIFUNCS_API_KEY). The missing declaration in the registry metadata is an inconsistency and reduces trust.
Instruction Scope
SKILL.md instructs the agent to run the bundled Python scripts and to set UNIFUNCS_API_KEY; the scripts perform only network calls to the UniFuncs API, stream output to temp or user-specified files, and print results. These actions are within the stated 'deep search' scope. Two points to watch: (1) options like --push-to-share and --set-public can publish results to the service's share space, and (2) streaming writes/read stream-file contents to disk (temp or user-specified paths). Both behaviors can expose query contents if the user provides sensitive data.
Install Mechanism
No install spec is present (instruction-only with included scripts). That is lower risk because nothing is downloaded or executed from an external arbitrary URL on install. The code itself is bundled and uses only the Python standard library (urllib, tempfile, etc.).
Credentials
The code and SKILL.md require a UNIFUNCS_API_KEY, but registry metadata lists no required env vars or primary credential — this mismatch is a notable red flag. Aside from that API key, no other credentials are requested. Because queries (and any sensitive content included in them) will be transmitted to a third-party API, the API key should be declared by the publisher, and users should ensure the key has appropriate scope/limits.
Persistence & Privilege
The skill is not marked always:true and does not request permanent system privileges. It writes ephemeral stream files to temp or user-specified paths and does not modify other skills or global agent settings. Autonomous invocation is allowed (platform default) but not combined with other high-risk flags.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install unifuncs-deep-search - 安装完成后，直接呼叫该 Skill 的名称或使用
/unifuncs-deep-search触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v0.0.5
- Added compiled Python file: __pycache__/deep-search-report.cpython-313.pyc
- No changes to skill usage, documentation, or features.
- This release is functionally equivalent to the previous version except for the addition of the bytecode file.
v0.0.3
- Major update: Deep Search skill restructured into three independent Python scripts for flexible synchronous and asynchronous operation.
- Added: Python scripts for task creation, querying, and synchronous report generation (`deep-search-create-task.py`, `deep-search-query-task.py`, `deep-search-report.py`).
- Removed: Legacy API documentation and shell script.
- Improved: SKILL.md rewritten in English with clearer usage scenarios, user instructions, and full CLI option reference.
- Enhanced guidance on when and how to use deep search, as well as on task management and timeout expectations.
v0.0.1
Initial release of unifuncs-deep-search skill.
- Enables deep, fast, and comprehensive information search via UniFuncs API.
- Supports using API Key for authenticated requests.
- Provides Bash (curl) usage examples and instructions.
- Allows configurable parameters for search queries.
- Documentation links included for further reference.
元数据
常见问题
UniFuncs Deep Search 是什么？
Use UniFuncs Deep Search API for fast, comprehensive information gathering. Use this skill when users ask for deep search, broad investigation, or in-depth t... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 1134 次。
如何安装 UniFuncs Deep Search？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install unifuncs-deep-search」即可一键安装，无需额外配置。
UniFuncs Deep Search 是免费的吗？
是的，UniFuncs Deep Search 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
UniFuncs Deep Search 支持哪些平台？
UniFuncs Deep Search 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 UniFuncs Deep Search？
由 UniFuncs（@vinlic）开发并维护，当前版本 v0.0.5。
推荐 Skills