← 返回 Skills 市场
14435
总下载
8
收藏
104
当前安装
43
版本数
在 OpenClaw 中安装
/install cellcog
功能描述
Any-to-any AI sub-agent — research, images, video, audio, music, podcasts, avatars, voice cloning, documents, spreadsheets, dashboards, 3D models, diagrams,...
安全使用建议
Install only if you are comfortable using CellCog's cloud service, setting a CellCog API key, spending service credits, uploading only files you intentionally wrap in SHOW_FILE tags, and allowing generated files to download to requested or default CellCog paths. Avoid tagging secrets, private keys, .env files, SSH keys, confidential documents, or files you do not want sent to CellCog.
功能分析
Type: OpenClaw Skill
Name: cellcog
Version: 2.0.15
The 'cellcog' skill (SKILL.md) provides an interface for an external AI service with high-risk capabilities, including the ability to upload any local file via absolute paths and download files to the local filesystem. While the documentation contains explicit warnings against sharing sensitive information like credentials or SSH keys, the broad file and network access are inherently risky. The mention of a background daemon for task monitoring and the use of future-dated metadata (April 2026) in `_meta.json` and `SKILL.md` are additional indicators that warrant a cautious classification despite the lack of clear malicious intent.
能力评估
Purpose & Capability
The broad any-to-any AI sub-agent capability is real, including research, media generation, documents, spreadsheets, code, and multimodal files, but it matches the stated purpose of delegating work to CellCog rather than showing unrelated behavior.
Instruction Scope
Instructions clearly disclose that files inside SHOW_FILE tags are uploaded and warn against credentials, private keys, .env files, SSH keys, and other sensitive material. The instruction to always print the full result message is a privacy caution because returned text can include paths, credits, follow-up instructions, or content derived from uploaded files.
Install Mechanism
The artifact is a single non-executable SKILL.md. It tells users to install the CellCog Python SDK with pip if needed and requires CELLCOG_API_KEY, but contains no bundled scripts, binaries, startup hooks, or privileged installer.
Credentials
Cloud API use, an API key, intentional file uploads, account credit usage, and generated-file downloads are proportionate for this service, but users should treat them as sensitive external data sharing.
Persistence & Privilege
OpenClaw notify mode mentions a background WebSocket daemon for completion notifications and resilient delivery, but this is disclosed and tied to submitted CellCog tasks rather than hidden persistence or privilege escalation.
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 cellcog - 安装完成后，直接呼叫该 Skill 的名称或使用
/cellcog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
v2.0.15
- Documentation clarifies that anything wrapped in <SHOW_FILE> tags is uploaded to CellCog; warns not to include credentials, private keys, or sensitive material as SHOW_FILE files.
- Added a comparison to reference image attachments in Nano Banana to explain <SHOW_FILE> usage.
- Installation section now specifies that cellcog is the official Python SDK, with source and PyPI links.
- Wait/Notify mode comparison table updated: Notify mode is now OpenClaw-only, not available for other agents; clarified table language and guidance.
- Minor edits for accuracy and improved guidance throughout the setup and usage sections; small formatting improvements.
v2.0.14
- Added support for new modalities: avatars and voice cloning in the SKILL description.
- Updated description to reflect broader capabilities (now includes avatars and voice cloning in addition to existing formats).
- No code or API changes—documentation/feature list update only.
v2.0.13
- Expanded description to emphasize agent-to-agent protocol and multi-step iteration for high accuracy.
- Updated and clarified introductory sections to stress support for all output types, not just code.
- Improved highlights of CellCog's DeepResearch Bench #1 ranking and modal capabilities.
- Cleaned up and streamlined the documentation for easier understanding without removing key usage instructions.
- Clarified that CellCog ensures all your work gets done, in any modality.
v2.0.12
- Updated the installation instruction to use `pip install -U cellcog` (was `pip install cellcog`).
- No functional or API changes introduced in this update.
- Documentation remains focused on setup, usage, and agent integration.
v2.0.11
- SKILL.md significantly shortened and streamlined.
- Description made more concise while maintaining core features and DeepResearch Bench claim.
- Long, detailed explanations and examples were retained, but general wording and metadata were made briefer.
- Content after the "Response Shape" heading was cut off for brevity.
- No changes to skill logic or API; documentation update only.
v2.0.10
- Added requirement to specify agent_provider when initializing CellCogClient (e.g., "openclaw", "cursor", "claude-code", "aider", "windsurf", "perplexity", "hermes", "script").
- Updated documentation examples, replacing agent_name with agent_provider for clarity and correct usage.
- Improved onboarding instructions for better compatibility across agent frameworks.
v2.0.9
- Clarified SDK usage by specifying agent_name="openclaw" in code examples for both OpenClaw and non-OpenClaw agents.
- Updated description to remove "deep research" to improve clarity.
- No other functionality or interface changes.
v2.0.8
- Updated description to highlight deep research as a primary use case.
- Minor text edits for clarity and emphasis in SKILL.md.
- No functional or code-level changes; documentation only.
v2.0.7
cellcog 2.0.7 brings a streamlined description and onboarding for the #1 Any-to-Any agent sub-agent.
- Updated SKILL.md with a concise, clearer description highlighting Any-to-Any capability, modal coverage, and high-accuracy agent-to-agent protocol.
- Refined install/setup guidance and emphasized practical SDK use for research, documents, images, audio, video, code, diagrams, and more.
- Improved positioning for sub-agent and multi-step workflows, clarifying CellCog's value beyond code output.
- No code or API changes in this version; documentation only.
v2.0.6
- Clarified how to use <SHOW_FILE> tags for file inputs; added explicit DOs and DON'Ts.
- Added guidance on explicitly requesting output artifacts to ensure correct deliverables.
- Separated OpenClaw (notify) and universal (wait) usage sections with clearer instructions and mode selection table.
- Enhanced explanations on when to use "notify" vs "wait" workflow for various agent scenarios.
- Improved warnings on always printing the full response message.
- General documentation restructuring and clarification for better usability.
v2.0.5
CellCog 2.0.5 – Revised API Usage & Docs
- Updated SKILL.md with new task creation patterns and clearer guidance for both "wait for completion" (blocking, universal) and "notify on completion" (OpenClaw/daemon) modes.
- Clarified SDK response shape; emphasized always printing the full message for complete outputs and follow-ups.
- Improved timeout and resumption handling instructions.
- Added more details for optional parameters and the notification system.
- Removed SKILL_original.md (now redundant).
v2.0.4
- Documentation updated in SKILL.md.
- No functional or API changes; content only.
- Existing examples and instructions retained.
- No impact to skill operation or user integrations.
v2.0.3
- Documentation formatting updated in SKILL.md to improve readability.
- No functional or code changes; content remains the same.
- No new features or bug fixes included.
v2.0.2
- Updated DeepResearch Bench reference from February 2026 to April 2026 for all leaderboard mentions.
- Clarified and simplified instructions for file path usage and output requests.
- Streamlined credit usage explanation, noting orchestration of 21+ foundation models and removal of prior credit estimate discussion.
- Expanded documentation for task creation, including all optional parameters and required fields.
- Added a detailed "Chat Modes" section describing available modes, their use cases, speed, and credit requirements.
- Improved guidance and examples for session key usage and workflow instructions.
v2.0.1
- Documentation in SKILL.md has been updated.
- No changes to core functionality or interface.
v2.0.0
cellcog 2.0.0
- Added `homepage` field with the project website.
- Refactored and expanded metadata: clarified supported operating systems and environment variables, and added structured install instructions.
- Improved standardization and clarity in the SKILL.md header for better integration and discoverability.
- No changes to core usage documentation or API content.
v1.0.26
cellcog 1.0.26
- Expanded API reference for create_chat(): now documents optional project_id and agent_role_id parameters for advanced workflows.
- No changes to code or behavior; update improves user guidance and advanced usage documentation.
v1.0.25
- Removed the "Typical Credit Costs" table; documentation now explicitly states that credit usage estimates are not provided, with an explanation of why.
- Updated the "Quick Start" section to clarify how credit consumption works and to set expectations for users regarding variability.
- No changes to API or functionality; documentation update only.
v1.0.24
- Documentation updated in SKILL.md; no functional code changes.
- Ensures that instructions, usage examples, and explanations remain consistent and clear for all users.
v1.0.23
cellcog 1.0.23
- Added documentation for wait_for_completion(), enabling agents to block until a CellCog task finishes.
- Clarified guidance on using wait_for_completion() in automated workflows like cron jobs or pipelines.
- Minor text and formatting improvements in usage instructions.
- No code changes; update affects documentation only.
元数据
常见问题
Cellcog 是什么？
Any-to-any AI sub-agent — research, images, video, audio, music, podcasts, avatars, voice cloning, documents, spreadsheets, dashboards, 3D models, diagrams,... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 14435 次。
如何安装 Cellcog？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install cellcog」即可一键安装，无需额外配置。
Cellcog 是免费的吗？
是的，Cellcog 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Cellcog 支持哪些平台？
Cellcog 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（darwin, linux, windows）。
谁开发了 Cellcog？
由 CellCog（@nitishgargiitd）开发并维护，当前版本 v2.0.15。
推荐 Skills