Skip to main content

Provider Overview

Memory Swarm supports 5 provider types. Each provider has different capabilities, search methods, and requirements.

ByteRover

ByteRover’s context tree is always available as a swarm provider. It contributes the same BM25 search that powers brv query, without the LLM synthesis step. Capabilities:
  • Keyword search (BM25) across all curated knowledge files
  • Read-only — writes go through brv curate separately
  • No API key or external dependency required
When results come from ByteRover: Results are labeled [byterover] and include the context tree file path. RRF weight: 1.0 (highest confidence — curated knowledge)

Obsidian

Searches your Obsidian vault using an in-process MiniSearch index (BM25). The index is built lazily on first query and rebuilt when vault files change. Capabilities:
  • Keyword search across all .md files in the vault
  • Wikilink-aware — follows [[links]] for richer indexing
  • Graph traversal support for connected notes
  • Read-only in swarm mode
Configuration:
Requirements: An existing Obsidian vault directory (must contain .obsidian/ subdirectory). RRF weight: 0.85

Memory Wiki (OpenClaw)

Searches an OpenClaw Memory Wiki vault — a structured knowledge base with pages organized by type (entities, concepts, syntheses, sources). Uses in-process MiniSearch for fast keyword search. Capabilities:
  • Keyword search across wiki pages (entities, concepts, syntheses, sources)
  • Freshness boost — recently updated pages score 1.2x higher
  • Kind boost — entities (1.4x) and concepts (1.3x) are prioritized over sources
  • Write support — creates new concept or entity pages with OpenClaw frontmatter
Configuration:
Write behavior: Creates pages in concepts/ or entities/ with OpenClaw-compatible frontmatter:
Requirements: An OpenClaw wiki vault. Set up with openclaw wiki init or configure the path to an existing vault. RRF weight: 0.9

GBrain

Queries a GBrain knowledge brain via CLI subprocess. GBrain supports three search modes with different performance and cost tradeoffs. Search Modes: Configuration:
Requirements:
  • GBrain repo or global installation (bun add -g gbrain)
  • For hybrid/vector modes: OpenAI API key for embeddings
  • If running from source: Bun runtime
GBrain resolution order:
  1. Explicit gbrain_bin_path in config
  2. gbrain in system PATH
  3. src/cli.ts in repo_path
  4. ../gbrain/src/cli.ts relative to workspace
  5. Fallback — try gbrain anyway (fails at runtime if not found)
RRF weight: 0.85

Local Markdown

Indexes one or more folders of plain markdown files. Each folder is registered as a separate sub-provider (e.g., local-markdown:notes, local-markdown:docs), allowing fine-grained routing and write control. Capabilities:
  • Keyword search (BM25) with MiniSearch
  • Wikilink following for cross-file references
  • Write support (when read_only: false)
  • Multiple independent folders
Configuration:
Write behavior: New files are created as timestamped markdown files in the target folder (e.g., note-1776052527043.md). RRF weight: 0.8

Provider Weights

Each provider has an RRF weight that reflects confidence in its result quality. Higher weights mean the provider’s top results are more likely to appear at the top of the fused ranking.