Namespace: query — Search & Retrieval

  • Purpose: Search memories.
  • When to use: Retrieve relevant information by query.
  • What it does: Full-text/semantic search with optional filters.
  • Step-by-step:
    1. Provide --agent-id.
    2. Provide --query (optional for list-like behavior).
    3. Apply filters if needed.
  • Usage example: 
    afs query search --agent-id myagent --query "auth bug" --limit 10
  • Expected output: JSON array of matching memories.

afs query recall

  • Purpose: Graph-walking reflective recall.
  • When to use: Context reconstruction via relationships.
  • What it does: Walks memory graph and ranks results.
  • Usage example: 
    afs query recall --agent-id myagent --query "payment issue" --max-depth 2 --limit 10
  • Expected output: JSON with memory_id, relevance, and path.

afs query context

  • Purpose: Build LLM-ready context block.
  • When to use: Prompt injection or summarization.
  • What it does: Packs memory snippets into a character budget.
  • Usage example: 
    afs query context --agent-id myagent --query "security summary" --max-chars 2000
  • Expected output: JSON context object.

afs query recent

  • Purpose: List recent memories.
  • When to use: Quick review of latest items.
  • Usage example: 
    afs query recent --agent-id myagent --limit 10

afs query priority

  • Purpose: List high-importance memories.
  • When to use: Focus on critical memories.
  • Usage example: 
    afs query priority --agent-id myagent --min-importance 0.8 --limit 10

afs query inspect

  • Purpose: Detailed memory report for an agent.
  • When to use: Diagnostics and audits.
  • Usage example: 
    afs query inspect --agent-id myagent

afs query stats

  • Purpose: Memory statistics summary.
  • When to use: Monitoring.
  • Usage example: 
    afs query stats --agent-id myagent