hebbs - Cognitive memory and retrieval for AI agents
Indexes files and memories into propositions and entity graphs, then retrieves, reflects on, and resolves contradictions in stored knowledge.
Tags
Updated: 2026-10-08Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Index files into propositions
- Build entity graphs
- Store non-file memories
- Retrieve across weighted dimensions
- Generate consolidated insights
- Detect and resolve contradictions
- Check vault health
Inputs
- Files
- Memory facts
- Retrieval queries
- Project directory
- LLM provider configuration
- API credentials
Outputs
- Indexed memories
- Proposition and entity graphs
- Retrieval results
- Generated insights
- Resolved contradictions
- Vault health status
- Interactive memory graph
Requirements
- hebbs command-line binary
- Configured LLM provider
- Supported provider: OpenAI, Anthropic, Gemini, or Ollama
- API key for cloud providers