← 返回 Skills 市场
ppop0uuiu

local-model-router

作者 ppop0uuiu · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
202
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install local-model-router
功能描述
自动生成科技新闻摘要。从多个来源(RSS、Twitter、GitHub、Web Search)抓取科技新闻,整合后生成摘要。
使用说明 (SKILL.md)

Tech News Digest

自动科技新闻摘要生成器。

功能

  • 从 6 个来源并行抓取新闻:RSS、Twitter、GitHub、Web Search 等
  • 自动去重、评分、排序
  • 支持 Discord、Email、PDF 模板输出

使用方法

配置

# 复制默认配置
mkdir -p workspace/config
cp config/defaults/sources.json workspace/config/
cp config/defaults/topics.json workspace/config/

环境变量(可选)

  • TWITTERAPI_IO_KEY - Twitter API
  • X_BEARER_TOKEN - X Twitter
  • TAVILY_API_KEY - 搜索 API
  • BRAVE_API_KEY - Brave Search
  • GITHUB_TOKEN - GitHub

生成摘要

python3 scripts/run-pipeline.py \
  --defaults config/defaults \
  --config workspace/config \
  --hours 48 \
  --output /tmp/td-merged.json

注意

  • 需要 Python 3
  • 需要安装依赖:pip install -r requirements.txt
  • 免费 API 额度有限
安全使用建议
This package appears to be a simple RSS news summarizer, but there are clear inconsistencies you should consider before installing: the skill's registry name ('local-model-router') doesn't match its content; SKILL.md instructs running a non-existent 'scripts/run-pipeline.py'; it claims integration with Twitter/GitHub/Brave/Tavily and output channels (Discord/Email/PDF) that are not implemented in the included scripts; and the translator library used in code (deep_translator) is not listed in requirements.txt. Do not provide API keys unless you can confirm the code actually uses them. If you still want to use it, review the code locally, add the missing dependency (or remove unused env var prompts), and fix or implement the missing run-pipeline script or update SKILL.md to match the real behavior.
功能分析
Type: OpenClaw Skill Name: local-model-router Version: 1.0.1 The skill bundle is a legitimate tech news aggregator designed to fetch RSS feeds from major AI companies (OpenAI, Google, Microsoft) and generate translated summaries. The provided Python scripts (daily-digest.py and fetch-news.py) use standard libraries like feedparser and deep-translator to process news data and save it to a local workspace. No evidence of data exfiltration, malicious execution, or prompt injection was found; the code logic is transparent and aligns perfectly with the functionality described in SKILL.md.
能力评估
Purpose & Capability
Metadata name is 'local-model-router' while the description, SKILL.md, and code implement a tech-news digest — name/purpose mismatch. The README claims outputs like Discord/Email/PDF, but the included scripts only fetch RSS feeds and write local JSON/markdown/text files. Several optional API keys are listed in SKILL.md (Twitter, Brave, TAVILY, GitHub) but the provided Python scripts only read RSS feeds and do not call those services.
Instruction Scope
SKILL.md instructs creating a workspace config and running 'python3 scripts/run-pipeline.py' to produce output; no scripts/run-pipeline.py exists in the bundle. SKILL.md also promises parallel fetch from multiple sources (RSS, Twitter, GitHub, Web Search) and various output channels, but the actual scripts (fetch-news.py and daily-digest.py) only parse RSS feeds, translate text (optional), and save local files. The SKILL.md lists environment variables that the code does not read. This discrepancy grants the agent broad discretion without implementation to justify it.
Install Mechanism
No install spec (instruction-only), which minimizes installer risk. SKILL.md recommends 'pip install -r requirements.txt' and a requirements.txt is present (feedparser, requests, python-dateutil). However the code optionally imports deep_translator (GoogleTranslator) but deep_translator is not in requirements.txt, indicating incomplete dependency specification.
Credentials
SKILL.md lists multiple API keys as optional (TWITTERAPI_IO_KEY, X_BEARER_TOKEN, TAVILY_API_KEY, BRAVE_API_KEY, GITHUB_TOKEN) but none of the included scripts access these environment variables. Asking for many credentials unrelated to the provided implementation is disproportionate and could confuse users into supplying unnecessary secrets.
Persistence & Privilege
Skill is not always-enabled, does not request system-wide config changes, and does not declare required config paths or credentials. Scripts write files under scripts/workspace which is local and expected for this use case.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install local-model-router
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /local-model-router 触发
  4. 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
v1.0.1
根据任务需要智能选择本地模型
v1.0.0
Initial release of tech-news-digest skill. - Automatically generates tech news digests from multiple sources (RSS, Twitter, GitHub, Web Search). - Deduplication, scoring, and ranking of news items. - Outputs available as Discord, Email, and PDF templates. - Customizable via config files and supports multiple API keys for enhanced coverage. - Requires Python 3 and dependency installation.
元数据
Slug local-model-router
版本 1.0.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 2
常见问题

local-model-router 是什么?

自动生成科技新闻摘要。从多个来源(RSS、Twitter、GitHub、Web Search)抓取科技新闻,整合后生成摘要。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 202 次。

如何安装 local-model-router?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install local-model-router」即可一键安装,无需额外配置。

local-model-router 是免费的吗?

是的,local-model-router 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

local-model-router 支持哪些平台?

local-model-router 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 local-model-router?

由 ppop0uuiu(@ppop0uuiu)开发并维护,当前版本 v1.0.1。

💬 留言讨论