← 返回 Skills 市场
Ollama Local
作者
Timverhoogt
· GitHub ↗
· v1.1.0
6563
总下载
11
收藏
59
当前安装
2
版本数
在 OpenClaw 中安装
/install ollama-local
功能描述
Manage and use local Ollama models. Use for model management (list/pull/remove), chat/completions, embeddings, and tool-use with local LLMs. Covers OpenClaw sub-agent integration and model selection guidance.
安全使用建议
Install this if you intend to manage or use Ollama from OpenClaw. Keep OLLAMA_HOST on localhost for sensitive prompts, use only trusted remote Ollama servers, and double-check the target host before running pull, rm, or sub-agent workflows because those actions can consume disk/network resources or remove model files.
功能分析
Type: OpenClaw Skill
Name: ollama-local
Version: 1.1.0
The skill bundle is designed to manage and interact with local Ollama models. All Python scripts (`scripts/ollama.py`, `scripts/ollama_tools.py`) use standard `urllib.request` to communicate with a user-configured Ollama host, which is essential for their stated purpose. Crucially, the `run_code` tool within `scripts/ollama_tools.py` is explicitly a mock implementation and does not execute arbitrary code, preventing a significant potential vulnerability. The markdown files (`SKILL.md`, `references/models.md`) provide clear instructions and documentation without any evidence of prompt injection attempts or malicious directives for the agent. There is no indication of data exfiltration, unauthorized execution, or persistence mechanisms.
能力评估
Purpose & Capability
The documentation and scripts consistently support local or user-configured Ollama model management, chat, generation, embeddings, and tool-use examples.
Instruction Scope
Commands are user-directed examples; the remote OLLAMA_HOST example and direct API calls could warn more clearly that prompts and embeddings go to the configured server.
Install Mechanism
The bundle contains markdown and Python helper scripts only, with no installer, autostart behavior, dependency installation, or privilege escalation mechanism.
Credentials
Reading OLLAMA_HOST and making HTTP requests to that host is proportionate for Ollama integration, but users should treat remote hosts as data recipients.
Persistence & Privilege
No persistence or credential harvesting is present; pull and remove operations can change the configured Ollama server's model inventory when explicitly run.
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 ollama-local - 安装完成后，直接呼叫该 Skill 的名称或使用
/ollama-local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
v1.1.0
Generalized for any Ollama setup - host now configurable via OLLAMA_HOST env var (defaults to localhost:11434)
v1.0.0
Initial release - CLI for model management, chat/completions, embeddings, and tool-use with local Ollama models
元数据
常见问题
Ollama Local 是什么？
Manage and use local Ollama models. Use for model management (list/pull/remove), chat/completions, embeddings, and tool-use with local LLMs. Covers OpenClaw sub-agent integration and model selection guidance. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 6563 次。
如何安装 Ollama Local？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install ollama-local」即可一键安装，无需额外配置。
Ollama Local 是免费的吗？
是的，Ollama Local 完全免费（开源免费），可自由下载、安装和使用。
Ollama Local 支持哪些平台？
Ollama Local 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 Ollama Local？
由 Timverhoogt（@timverhoogt）开发并维护，当前版本 v1.1.0。
推荐 Skills