mcpField
C
chat-format

chat-format

github.com/ruvnetAgent workflows80.0k742 installsUpdated today
Repository

About chat-format

chat-format is a skill for formatting prompts for multiple LLM providers and preparing context retrieval with HNSW. It covers provider-specific chat formats, document indexing, query routing, and provider status checks.

Key features

  1. 01Format chats for multiple providers
  2. 02Create HNSW context indexes
  3. 03Add documents to an index
  4. 04Route queries to relevant context
  5. 05Check provider availability

Use cases

  1. 01Prepare prompts for Claude, GPT, Gemini, or Ollama
  2. 02Build an HNSW-based RAG pipeline
  3. 03Retrieve indexed context for a query
  4. 04Check which providers are available

MCP support

Uses Ruflo MCP tools to format chats, create and populate HNSW indexes, route retrieval queries, and check provider status.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/ruvnet/ruflo --skill chat-format

After installing, type/chat-formatin Claude Code to confirm it loaded.

Details

From the project README

Chat Format

Format prompts for multi-provider LLM inference with context retrieval.

When to use

When preparing prompts for different LLM providers (Claude, GPT, Gemini, Ollama) or building RAG pipelines with HNSW-powered context retrieval.

Steps

  1. Format chat — call mcp__plugin_ruflo-core_ruflo__ruvllm_chat_format with messages and target provider
  2. Create HNSW index — call mcp__plugin_ruflo-core_ruflo__ruvllm_hnsw_create for context retrieval
  3. Add documents — call mcp__plugin_ruflo-core_ruflo__ruvllm_hnsw_add to index documents
  4. Route query — call mcp__plugin_ruflo-core_ruflo__ruvllm_hnsw_route to find relevant context
  5. Check status — call mcp__plugin_ruflo-core_ruflo__ruvllm_status for provider availability

Supported providers

  • Anthropic (Claude) — native format
  • OpenAI (GPT) — chat completion format
  • Google (Gemini) — generative AI format
  • Ollama — local model format
  • Cohere — generate/chat format

Comments

Markdown supported