← 返回 Skills 市场

Pywayne Llm Chat Ollama Gradio

Name: Pywayne Llm Chat Ollama Gradio
Author: wangyendt

作者 wangyendt · GitHub ↗ · v0.1.0

cross-platform ⚠ suspicious

639

总下载

当前安装

版本数

在 OpenClaw 中安装

/install chat-ollama-gradio

功能描述

Gradio-based chat interface for Ollama with multi-session management. Use when working with pywayne.llm.chat_ollama_gradio module to create a web-based chat...

使用说明 (SKILL.md)

Pywayne Chat Ollama Gradio

This module provides a Gradio-based web chat interface for Ollama models with multi-session support.

Quick Start

from pywayne.llm.chat_ollama_gradio import OllamaChatGradio

# Create and launch chat interface
app = OllamaChatGradio(
    base_url="http://localhost:11434/v1",
    server_port=7870
)
app.launch()

Configuration

Parameter	Default	Description
`base_url`	`"http://localhost:11434/v1"`	Ollama API base URL
`server_name`	`"0.0.0.0"`	Server host name
`server_port`	`7870`	Server port
`root_path`	`""`	Root path for reverse proxy
`api_key`	`"ollama"`	API key (for Ollama compatibility)

Model Discovery

Automatically discovers available Ollama models by running ollama list:

Excludes models with 'embed' in the name
Falls back to qwen2.5:0.5b if no models found

Session Management

Creating New Sessions

# UI method: Click "新建会话" button
new_chat_id, new_history, new_choices = app.create_new_chat()

Switching Sessions

# UI method: Select from "历史会话" radio list
history = app.switch_chat(selected_chat_id)

API Reference

OllamaChatGradio

Method	Description
`get_ollama_models()`	Get list of available Ollama models
`init_chat_manager()`	Initialize ChatManager instance
`create_new_chat()`	Create new chat session, returns (chat_id, history, radio_update)
`switch_chat(chat_id)`	Switch to specified chat session
`format_history(history)`	Format history for display
`chat(message, history, model_name)`	Process chat message with streaming
`create_demo()`	Create Gradio interface
`launch()`	Launch Gradio server

UI Components

Component	Description
`chatbot`	Main chat display area
`msg`	Message input textbox
`model_dropdown`	Model selection dropdown
`chat_id_text`	Current session ID (read-only)
`new_chat_btn`	Button to create new session
`chat_history_list`	Radio list for session switching

Requirements

gradio - Web UI framework
pywayne.llm.chat_bot - ChatManager and LLMConfig
ollama CLI - Must be installed and accessible

Notes

Uses ChatManager for multi-session support
Streaming responses update UI in real-time
Session history persists in memory (not persistent)
Requires Ollama to be running before launching

安全使用建议

What to check before installing or using this skill: - The SKILL.md requires the 'ollama' CLI, 'gradio', and the 'pywayne.llm.chat_bot' module, but the registry metadata lists none of these — verify you have these dependencies from trusted sources before running anything. - The skill will call 'ollama list' (a local CLI command) and launch a local web server (default port 7870). Run it in an isolated environment (container/VM) if you are unsure of the origin. - Inspect the actual pywayne.llm.chat_ollama_gradio and pywayne.llm.chat_bot code (or obtain them from a reputable source) to confirm there are no unexpected network calls, file reads, or credential handling. - Confirm your Ollama CLI was installed from a trusted release (official site/GitHub) so that 'ollama list' isn't an unexpected binary. - If you need higher confidence: ask the publisher for the source repository, a checksum for any required packages/binaries, or a clear install spec. If those cannot be provided, treat the skill as untrusted and run it in isolation.

功能分析

Type: OpenClaw Skill Name: chat-ollama-gradio Version: 0.1.0 The skill is classified as suspicious due to its explicit reliance on executing shell commands (`ollama list`) for model discovery and binding a web server to all network interfaces (`0.0.0.0`) by default, as described in `SKILL.md`. While these capabilities are plausibly needed for the skill's stated purpose of providing a Gradio-based chat interface for Ollama, they represent significant attack surfaces. Without access to the underlying Python code, it's impossible to verify proper input sanitization for shell commands or secure configuration of the web server, making it a risky capability rather than clearly benign.

能力评估

⚠ Purpose & Capability

The skill's stated purpose (Gradio UI for Ollama) matches the instructions, but the registry metadata lists no required binaries or environment variables while the SKILL.md explicitly requires the 'ollama' CLI, the 'gradio' package, and the 'pywayne.llm.chat_bot' module. That omission is an inconsistency between what the skill says it needs and what it actually instructs the agent to use.

✓ Instruction Scope

The instructions stay within the stated purpose: they import a pywayne module, launch a Gradio server, and call 'ollama list' to discover local models. The SKILL.md does not instruct reading unrelated system files, exfiltrating data, or contacting external endpoints beyond the local Ollama API and a locally hosted Gradio UI. It does, however, expect a local service and CLI to be present.

✓ Install Mechanism

This is an instruction-only skill with no install spec, so nothing will be downloaded or written by the skill itself. That lowers install risk, but the lack of an install spec combined with unmet runtime dependencies (ollama CLI, gradio, pywayne package) means a user may have to install components manually — verify sources before doing so.

✓ Credentials

The skill does not request environment variables or credentials in metadata. The SKILL.md mentions an 'api_key' parameter for Ollama compatibility but does not require secrets or other unrelated credentials. No disproportionate credential access is requested.

✓ Persistence & Privilege

The skill does not request always: true and is user-invocable only. It describes running a local Gradio server (default port 7870) and keeping session history in memory; it does not indicate modifying other skills or system-wide settings.

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 chat-ollama-gradio
安装完成后，直接呼叫该 Skill 的名称或使用 /chat-ollama-gradio 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

v0.1.0

- Initial release of pywayne-llm-chat-ollama-gradio: Gradio chat UI for Ollama models with multi-session management. - Supports model auto-discovery, excluding 'embed' models and providing a fallback option. - Includes chat session creation, switching, and management via integrated ChatManager. - Offers a configurable interface for model selection, API settings, and server options. - Streaming responses and chat history handled in real-time within the web UI.

元数据

Slug chat-ollama-gradio

版本 0.1.0

许可证 —

累计安装 0

当前安装数 0

历史版本数 1

常见问题