shared-pinecone-rag

What This Skill Does

The shared-pinecone-rag skill serves as the unified RAG (Retrieval-Augmented Generation) interface for OpenClaw agents. By utilizing a centralized Pinecone index located at /home/Mike/.openclaw/workspace/rag-pinecone-starter, it allows agents to maintain a long-term, queryable knowledge base without needing to duplicate data stores. This skill handles both the ingestion of markdown and text-based documentation into the 'pulse-rag' index and the retrieval of semantically relevant context for agent queries. When integrated with 'hybrid-db-health', it functions as a core component of a Persistent Memory stack, where this skill manages the data layer while the health module ensures reliability. It is the recommended standard for any workspace that requires shared knowledge across multiple agents.

Installation

To integrate this skill into your workspace, execute the following command in your terminal: clawhub install openclaw/skills/skills/cryptobro-man/shared-pinecone-rag

Ensure that the following environment prerequisites are met before initial use:

1. Verify that PINECONE_API_KEY is correctly configured inside rag-pinecone-starter/.env.
2. Ensure a functional Python virtual environment exists at rag-pinecone-starter/.venv.
3. Verify that your system has read/write permissions for the directory /home/Mike/.openclaw/workspace/rag-pinecone-starter/docs/.

Use Cases

• Technical Documentation Aggregation: Collect all project-related markdown docs in one place so agents can answer technical questions about project architecture.
• Persistent Context Sharing: Enable different agents to recall previous project decisions, meeting notes, or research findings stored in the 'default' namespace.
• Unified Knowledge Base: Prevent the fragmentation of data by forcing all agents to read from and write to the same source of truth rather than maintaining isolated, stale vector databases.

Example Prompts

1. "Query the shared Pinecone index to find out what our documentation says about the API rate limits for the gateway."
2. "I have just added a new set of design specification markdown files to the docs folder; please ingest these into the pulse-rag index."
3. "Retrieve the relevant context from the shared RAG memory regarding our security protocols for local database backups."

Tips & Limitations

• Shared Responsibility: This is a shared resource. Avoid deleting or overwriting documents unless you have workspace-wide authorization to prevent data loss for other agents.
• Maintenance: Periodically check the 'pulse-rag' index health. If the index size grows significantly, consider performing a cleanup of the 'docs' folder to ensure retrieval speed remains optimal.
• Naming Conventions: Use clear filenames in the docs directory, as the index will associate file names with the semantic chunks, making it easier to track the provenance of retrieved information.
• Security: Ensure the Pinecone API key is not exposed in public repositories, as it grants write access to your vector database.