Skip to content

Latest commit

 

History

History
174 lines (115 loc) · 6.15 KB

File metadata and controls

174 lines (115 loc) · 6.15 KB

MCP Tools

Back to README

All 7 tools available when Claude Code connects to the Recall MCP server (recall-memory).


memory_search

FTS5 keyword search across all memory tables. Use before asking the user to repeat anything.

Parameters

Name Type Required Default Description
query string yes Search query. FTS5 supports AND, OR, NOT, prefix*, "exact phrase"
project string no Filter results to a specific project name
table string no Restrict search to one table: messages, loa, decisions, learnings, breadcrumbs
limit number no 10 Maximum number of results to return

Returns: Array of matching records with table name, id, content, project, and snippet highlighting.

memory_search({ query: "kubernetes auth", project: "my-app", table: "decisions", limit: 10 })

memory_hybrid_search

Combined keyword + semantic search using Reciprocal Rank Fusion. Best for natural language queries. Falls back to keyword-only search if embeddings are unavailable.

Parameters

Name Type Required Default Description
query string yes Natural language search query
project string no Filter results to a specific project name
limit number no 10 Maximum number of results to return

Returns: Array of matching records ranked by fused keyword and semantic relevance scores.

memory_hybrid_search({ query: "how did we handle rate limiting", project: "my-app" })

memory_recall

Get recent context — LoA entries, decisions, and breadcrumbs. Good for orienting at the start of a session.

Parameters

Name Type Required Default Description
limit number no 5 Number of recent entries to return per category
project string no Filter results to a specific project name

Returns: Recent records grouped by category: Library of Alexandria entries, decisions, and breadcrumbs.

memory_recall({ limit: 5, project: "my-app" })

context_for_agent

Call this before spawning any agent via the Task tool. Uses hybrid search to find relevant memory context for the planned task, so the agent starts with relevant background.

Parameters

Name Type Required Default Description
agent_task string yes The task or prompt you plan to give the agent
project string no Current project name for filtering

Returns: Formatted context block containing relevant decisions, learnings, and breadcrumbs matched to the task description.

context_for_agent({ agent_task: "Refactor the auth middleware", project: "my-app" })

memory_add

Add structured records during a session. Use this to capture decisions, learnings, and work-in-progress breadcrumbs as they happen.

Parameters

Name Type Required Default Description
type string yes Record type: "decision", "learning", or "breadcrumb"
content string yes Main content
detail string no Additional detail — reasoning for decisions, solution steps for learnings
project string no Project name
tags string no Comma-separated tags (applies to learnings)

Returns: Confirmation with the new record's id and table.

memory_add({ type: "decision", content: "Use PostgreSQL over MySQL", detail: "Better JSON support and JSONB indexing" })
memory_add({ type: "learning", content: "bun:sqlite uses $param syntax", detail: "Not :param like better-sqlite3", tags: "bun,sqlite" })
memory_add({ type: "breadcrumb", content: "Auth refactor in progress — do not touch middleware until complete" })

memory_stats

Get database statistics including record counts per table and total database size. No parameters.

Returns: Record counts for each table (messages, decisions, learnings, breadcrumbs, loa, docs) and total database size on disk.

memory_stats()

loa_show

Show a full Library of Alexandria entry with its Fabric extract_wisdom content. LoA entries are curated knowledge summaries extracted from session conversations.

Parameters

Name Type Required Default Description
id number yes LoA entry ID (use memory_search or memory_recall to find IDs)

Returns: Full LoA record including title, summary, insights, quotes, and any Fabric-extracted wisdom.

loa_show({ id: 1 })

memory_dump

Flush the current conversation session into SQLite. Extracts messages, decisions, and learnings from the session transcript and persists them to the database. Use when the user says /dump.

Parameters

Name Type Required Default Description
title string yes Descriptive title for this session dump
project string no Override the auto-detected project name
skip_fabric boolean no true Skip Fabric processing (faster; uses a basic summary instead of extract_wisdom)

Returns: Summary of records imported: message count, decisions, learnings, and breadcrumbs extracted from the session.

memory_dump({ title: "Auth middleware refactor — JWT validation approach", project: "my-app" })

When to Use Each Tool

Scenario Tool
Starting a new session memory_recall
User asks about past work memory_search or memory_hybrid_search
Before spawning a sub-agent context_for_agent
Recording an architectural decision memory_add with type: "decision"
Capturing a technical insight memory_add with type: "learning"
Marking work-in-progress state memory_add with type: "breadcrumb"
End of session memory_dump
Viewing curated knowledge loa_show
Quick database health check memory_stats