← Back to Skills Marketplace
287
Downloads
0
Stars
0
Active Installs
13
Versions
Install in OpenClaw
/install summarize-all
Description
智能总结工具 - 自动检测内容类型、智能选择长度、缓存、关键点提取。支持网页、PDF、图片、音频、YouTube。
Usage Guidance
This package appears to do what it says: it summarizes content and sends material to whatever API endpoint and key you configure. Before installing or supplying an API key: (1) only set the endpoint to a provider you trust — any configured endpoint will receive your content; (2) be aware the tool stores the API key and history in ~/.config/summarize-all (ensure file permissions and remove sensitive data if needed); (3) optional features (server mode, webhooks) open network surfaces — only enable them intentionally and bind to safe interfaces; (4) optional helper binaries (pdftotext, yt-dlp, external CLIs) will be invoked if present — audit them or run in an isolated environment if you’re cautious; (5) if you want maximum safety, review scripts/summarize-all.py locally or run the tool in a sandbox/container before giving it access to sensitive files or credentials.
Capability Analysis
Type: OpenClaw Skill
Name: summarize-all
Version: 3.0.6
The 'summarize-all' skill is a comprehensive AI-powered summarization utility that processes web pages, PDFs, images, and multimedia content. The Python script `summarize-all.py` implements legitimate features including history management, keyword tracking with desktop notifications via `notify-send`, and a local HTTP server for programmatic access. While the tool handles sensitive API keys and performs outbound network requests to user-defined endpoints, its behavior is transparently documented in `SKILL.md` and aligns with its stated purpose. Subprocess calls for external tools like `pdftotext` and `pdftk` are used correctly for document processing, and no evidence of malicious intent or unauthorized data exfiltration was found.
Capability Assessment
Purpose & Capability
Name/description, SKILL.md and the Python script all describe a summarizer that supports webpages, PDFs, audio, images and YouTube. Required items (config file with endpoint/key, optional pdftotext/yt-dlp) match the stated functionality.
Instruction Scope
SKILL.md and the script instruct the agent to read local files (paths under user's control) and send content to the configured API endpoint. This is expected for a summarizer, but it does mean user content (including local files) will be transmitted to the configured API.
Install Mechanism
No install spec; this is an instruction-only skill with an included script. Nothing is downloaded or extracted automatically by the registry package itself.
Credentials
The skill declares no required environment variables or external credentials. It stores endpoint/key in ~/.config/summarize-all/config.json (user-supplied), which is proportionate to the purpose.
Persistence & Privilege
The tool writes config/history/cache files under ~/.config/summarize-all and can run an HTTP server and webhooks if the user starts those features. That behavior is expected for an API/server mode but expands network exposure if enabled.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install summarize-all - After installation, invoke the skill by name or use
/summarize-all - Provide required inputs per the skill's parameter spec and get structured output
Version History
v3.0.6
- Updated SKILL.md description: removed emoji and simplified to Chinese only.
- No feature or functionality changes; documentation only.
v3.0.5
- Minor documentation update: branding unified to "Summarize All Pro" and markdown formatting improved in SKILL.md.
- No changes to features or functionality.
- No code modifications.
v3.0.4
Major documentation update for Summarize-All v3.0.4:
- Expanded and clarified feature descriptions, including new options such as Markdown rendering and enhanced multi-model support.
- Added detailed sections for supported models (OpenAI, Anthropic, Google, OpenRouter, Chinese models).
- Included comprehensive usage examples for all command modes (interactive, batch, monitoring, translation, knowledge graph, etc.).
- Provided explicit configuration instructions and command references.
- Improved explanations for content types and summary length options.
- Updated privacy section with clearer guidance on endpoint and key management.
v3.0.3
- SKILL.md simplified and condensed: easier to read, less repetition, and concise descriptions.
- Features, options, commands, requirements, and supported content summarized in clearer tables and lists.
- Usage instructions streamlined for quick start and reference.
- Privacy section simplified; details clarified.
- Removed detailed explanations and examples, focusing on core information and brevity.
v3.0.2
- Major documentation update: SKILL.md rewritten for improved readability and English-language consistency.
- Feature/command descriptions, table headers, and option descriptions are now all in English.
- Some feature explanations are clarified and phrased for international users.
- Unnecessary duplicate details and transliterations were removed or simplified.
- No changes to the actual code or functionality—documentation only.
v3.0.1
**Major update with new features, model support, and command improvements:**
- Enhanced with auto content-type detection (news, papers, code, recipes, and more).
- Broadened model support: OpenAI (GPT-4o, 3.5), Anthropic, Google Gemini, OpenRouter, and major Chinese LLMs.
- Added advanced features: Q&A, translation, knowledge graph extraction, structured reports, quality scoring, auto re-summarization, and model comparison.
- Introduced API server, webhooks, batch processing, history/keyword/tag management, URL monitoring, and cache optimizations.
- Updated documentation with detailed commands, new interactive usage, configuration examples, and privacy notes.
v3.0.0
Summarize All 3.0.0 is a major upgrade introducing advanced smart features and new user commands.
- Added stream output, batch processing, search, keyword tracking, tag extraction, notifications, compare mode, and academic mode.
- Redesigned command syntax for batch, compare, academic, and search operations.
- Enhanced options including auto-length and auto-language detection, cache skipping, and verbose mode.
- Improved markdown rendering and typewriter effect for summaries.
- Expanded and reorganized documentation to highlight new features and usage patterns.
v1.4.2
- Added _meta.json file for improved metadata management.
- No changes to core functionality or documentation.
v1.4.0
- Removed the file scripts/summarize-all.sh.
- No changes to documentation or user-facing features.
- Maintenance update to reduce unused files.
v1.3.0
- Added a data privacy notice, highlighting that summarized content is sent to your configured API endpoint and users control which endpoint and API key to use.
- Updated requirements section to mention pdftotext and yt-dlp as optional tools for enhanced extraction.
v1.2.0
**Changelog for summarize-all v1.2.0**
- Refined model/API support: now focuses on OpenAI-compatible endpoints, simplifying configuration.
- Updated documentation to clarify requirements, usage, and supported content/methods.
- Generalized model support: 3rd party and Chinese models now require OpenAI-compatible API endpoints.
- Removed detailed provider/model lists in favor of concise guidance for proxies and compatibility.
- Streamlined instructions for setup and configuration.
v1.1.0
- Major update: Adds support for summarizing PDFs, images, audio files, and YouTube videos.
- Updated documentation to list all supported content types and provide usage examples.
- Expanded the skill description and instructions in SKILL.md for broader file and media compatibility.
- No external dependencies added; configuration remains unchanged.
v1.0.0
- Initial release of summarize-all.
- Summarize web pages, files, and text using any supported AI model.
- Compatible with major providers: OpenAI, Anthropic, Google Gemini, xAI Grok, OpenRouter, and a variety of Chinese and open-source models.
- Flexible configuration for endpoints, API keys, and models via command line.
- No external dependencies required.
- Supports custom proxy endpoints and summary length options.
Metadata
Frequently Asked Questions
What is Summarize All?
智能总结工具 - 自动检测内容类型、智能选择长度、缓存、关键点提取。支持网页、PDF、图片、音频、YouTube。 It is an AI Agent Skill for Claude Code / OpenClaw, with 287 downloads so far.
How do I install Summarize All?
Run "/install summarize-all" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Summarize All free?
Yes, Summarize All is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Summarize All support?
Summarize All is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Summarize All?
It is built and maintained by Qing WETH (@gu2003li); the current version is v3.0.6.
More Skills