← 返回 Skills 市场
🔌
Pangolinfo AI SERP: AI Mode Output + AI Overviews
作者
Pangolinfo
· GitHub ↗
· v2.0.2
· MIT-0
295
总下载
5
收藏
2
当前安装
7
版本数
在 OpenClaw 中安装
/install pangolinfo-ai-serp
功能描述
Search Google programmatically and extract structured SERP data, Google AI Overviews (SGE), citations, screenshots, and AI Mode follow-up results. Use when y...
安全使用建议
This skill appears internally consistent with its stated purpose. Before installing: 1) Prefer supplying PANGOLINFO_API_KEY over email+password to avoid storing your account password in the environment. 2) Be aware of optional caching: if you enable --cache-key or change CACHE_TO_DISK, a token may be written to ~/.pangolinfo_api_key; delete that file to revoke local persistence. 3) Confirm you trust the Pangolinfo service (https://pangolinfo.com and scrapeapi.pangolinfo.com) since network requests and your credentials go to that vendor. 4) Watch credit usage and rate limits (docs state requests consume credits on success). 5) If you need further assurance, review the remaining (truncated) parts of scripts/pangolinfo.py for any unexpected remote endpoints or behaviors—based on the provided files, no unrelated exfiltration or privileged actions were found.
功能分析
Type: OpenClaw Skill
Name: pangolinfo-ai-serp
Version: 2.0.2
The skill is a legitimate API client for the Pangolinfo SERP service, designed to extract Google search results and AI Overviews. The code in `scripts/pangolinfo.py` uses only Python standard libraries and communicates exclusively with the service's official endpoints (`scrapeapi.pangolinfo.com`). It includes a standard credential caching mechanism in the user's home directory (`~/.pangolinfo_api_key`) that is opt-in and uses secure file permissions (0o600). No evidence of malicious intent, unauthorized data exfiltration, or harmful prompt injection was found.
能力标签
能力评估
Purpose & Capability
Name/description ask for a Google SERP/SGE scraping capability and the package requires an API key (or email/password) and Python; the code calls scrapeapi.pangolinfo.com and auth endpoints that are consistent with that purpose.
Instruction Scope
SKILL.md instructs the agent to run the bundled Python script with expected flags and environment variables, describes outputs and error handling, and limits scope (explicit 'When NOT to use'). Instructions do not request unrelated files, system credentials, or network endpoints outside Pangolinfo.
Install Mechanism
Instruction-only with bundled scripts; no package downloads or installers. Requires only python3 on PATH. No high-risk archive downloads or unusual install behavior.
Credentials
Requires PANGOLINFO_API_KEY or PANGOLINFO_EMAIL + PANGOLINFO_PASSWORD which are directly used to authenticate to the Pangolinfo API. Number and type of env vars are proportional to the service integrated. Note: optional per-run caching to ~/.pangolinfo_api_key exists (disabled by default).
Persistence & Privilege
always:false and the skill does not request system-wide privileges. The only persistence option is the optional API key cache file under the user's home directory (CACHE_TO_DISK is False by default and caching is opt-in).
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 pangolinfo-ai-serp - 安装完成后，直接呼叫该 Skill 的名称或使用
/pangolinfo-ai-serp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
v2.0.2
Version 2.0.2: Major update — introduces Google SERP & AI Overview search functionality as a new, standalone skill.
- Now supports programmatic Google searching, AI Overview (SGE) extraction, citation retrieval, and SERP screenshots.
- Adds support for AI Mode with multi-turn follow-up questions for deeper query analysis.
- Amazon scraping features are removed and spun out to a separate skill ("Pangolinfo Amazon Scraper").
- New documentation and schema references for output structure.
- Designed for seamless use in agent frameworks and multistep workflows.
- Improved error handling for Google-specific use cases (rate limits, SGE availability, authorization).
v2.0.1
**Major documentation overhaul and onboarding clarity.**
- Completely revised skill documentation for easier onboarding and clearer capability overview.
- Added detailed user stories, common agent integrations, and multi-language usage notes.
- Updated Amazon scraping migration notice (now a separate skill).
- Streamlined prerequisites, quick-start prompts, and output field expectations.
- Expanded boundaries and error-handling guidance for users and agent developers.
- Now includes homepage, tags, and agent-facing feature summary in metadata.
v2.0.0
**Major update: Skill renamed/rebranded and all interfaces updated to "Pangolinfo".**
- Changed all environment variables and API references from `PANGOLIN_*` to `PANGOLINFO_*`.
- Main script renamed from `pangolin.py` to `pangolinfo.py`.
- Documentation updated to match new API key names, credential variables, and output/caching paths.
- All sample commands, error codes, and supporting text revised to Pangolinfo branding.
- Removed legacy Pangolin script; added updated Pangolinfo script.
v1.0.5
**Summary:** This release refines documentation, simplifies setup, and removes outdated references.
- Added a detailed first-time setup guide at `references/setup-guide.md`
- Removed older, redundant API reference and example files
- SKILL.md is now more concise with improved quickstart instructions and error/option tables
- "SERP Plus" (cheaper parser mode) and full CLI option coverage now documented
- All AI Agent response and error-handling rules are clarified and streamlined
v1.0.4
- Updated the skill description and documentation to emphasize the Google AI Mode search and structured JSON outputs with citations.
- Clarified that AI Mode is now the primary focus; SERP mode is available but not the main intent.
- Minor updates to naming and intent description; workflow, setup, and API usage remain unchanged.
- No functional/code changes in this version.
v1.0.2
Summary: Improved authentication flow, expanded documentation, and clearer usage intent.
• Authentication now prioritizes PANGOLIN_API_KEY (env) and cached ~/.pangolin_api_key before falling back to email/password login.
• Added --auth-only flow + first-time setup guide (cache API key once, then future runs need no env vars).
• Added macOS CERTIFICATE_VERIFY_FAILED troubleshooting section (SSL cert fix).
• Expanded intent-to-command mapping and “smart defaults” explanations for ai-mode / serp / screenshot / region / multi-turn follow-ups.
• Added references: output schema, error codes, API docs, plus JSON examples.
• Added scripts/self_test.sh for quick sanity checks.
tags：
• google-search
• serp
• ai-overview
• ai-mode
• web-search
• seo
• screenshot
• pangolinfo
• automation
• data-extraction
v1.0.0
Initial release with Google SERP + AI Overviews and multi-turn AI Mode.
- Programmatic Google search with structured JSON output via Pangolinfo APIs.
- Supports AI Overview extraction, AI Mode queries, multi-turn follow-ups, and optional page screenshots.
- Requires Pangolinfo auth via environment variables.
- Amazon scraping has moved to a separate skill.
元数据
常见问题
Pangolinfo AI SERP: AI Mode Output + AI Overviews 是什么？
Search Google programmatically and extract structured SERP data, Google AI Overviews (SGE), citations, screenshots, and AI Mode follow-up results. Use when y... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 295 次。
如何安装 Pangolinfo AI SERP: AI Mode Output + AI Overviews？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install pangolinfo-ai-serp」即可一键安装，无需额外配置。
Pangolinfo AI SERP: AI Mode Output + AI Overviews 是免费的吗？
是的，Pangolinfo AI SERP: AI Mode Output + AI Overviews 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Pangolinfo AI SERP: AI Mode Output + AI Overviews 支持哪些平台？
Pangolinfo AI SERP: AI Mode Output + AI Overviews 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（darwin, linux）。
谁开发了 Pangolinfo AI SERP: AI Mode Output + AI Overviews？
由 Pangolinfo（@pangolinfo）开发并维护，当前版本 v2.0.2。
推荐 Skills