菜单
Configure OpenRouter as the cloud-LLM backend for skills in this plugin. Use when a skill needs cloud LLM access and the user wants pay-as-you-go routing across Claude, GPT, Gemini, DeepSeek, Llama, Qwen without managing multiple provider keys.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/danielrosehill/Claude-Text-Corpus-Analysis-Plugin --skill setup-openrouter命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Assign each document in a corpus to one of N user-defined categories. Use when the user has a fixed taxonomy (e.g. 10-20 labels) and wants every note/document routed into exactly one (or top-k) of them. Supports zero-shot classifiers, local LLMs, and cloud LLMs with cost-aware batching.
Decide whether a corpus analysis task should use classical NLP, a local LLM, or a cloud LLM (OpenRouter) given corpus size, task complexity, and cost tolerance. Use first, before any other skill in this plugin, especially when the corpus is large (thousands+ of documents) or when an LLM pass could get expensive.
Correlate metadata (timestamps, tags, source, author) with content features (topics, entities, length, sentiment) to surface non-obvious patterns. Use when the user asks "does X correlate with Y in my corpus" or wants to discover relationships between when/where/who and what.
基于 SOC 职业分类
正在显示 SKILL.md
| name | setup-openrouter |
| description | Configure OpenRouter as the cloud-LLM backend for skills in this plugin. Use when a skill needs cloud LLM access and the user wants pay-as-you-go routing across Claude, GPT, Gemini, DeepSeek, Llama, Qwen without managing multiple provider keys. |
Single API key, access to ~every hosted LLM.
Key: check env for OPENROUTER_API_KEY. If absent, ask the user to set it (point at 1Password if they use op-vault). Never print the key.
Endpoint: https://openrouter.ai/api/v1 — set as OPENAI_BASE_URL with the OpenAI SDK.
Recommended model aliases (update when prices shift — check openrouter.ai/models):
| Tier | Model id (OpenRouter) | Good for |
|---|---|---|
| Ultra-cheap | deepseek/deepseek-chat, google/gemini-2.0-flash-001 | Bulk classification, labeling |
| Cheap | anthropic/claude-haiku-4-5, openai/gpt-4o-mini | Classification with higher quality, summarization |
| Premium | anthropic/claude-sonnet-4-6, anthropic/claude-opus-4-7, openai/gpt-4o | Taxonomy design, nuanced labeling, residue review |
| Embeddings | openai/text-embedding-3-small (direct OpenAI), or local sentence-transformers | Clustering, similarity |
Prices change — always fetch current pricing before large runs.
Test call: send a 1-token completion to verify routing + key.
Set safe defaults in a shared config:
Enable prompt caching for providers that support it (Anthropic, OpenAI) — huge savings on repeated system prompts in classification/taxonomy work.
choose-approach.max_spend_usd envelope; abort if projected exceed.logs/openrouter-calls.jsonl.OpenRouter forwards to the chosen provider. If the corpus is sensitive, prefer setup-local-llm.