Skip to content
Agentmart
Submit a tool

Memograph

Graph-based memory system for LLMs with knowledge graphs and semantic search.

Connector by Indhar01 · GitHub account verified by the MCP Registry

IndexedMITLocal (stdio)★ 4
Write a review

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/indhar01-memograph

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Graph-based memory system for LLMs with knowledge graphs and semantic search.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.

Tools
22 read · 11 write · 3 delete · 4 not yet classified
Secrets
2 (API keys, tokens or passwords)
Install scripts
Not scanned yet
Runs as
Python package

Tools by what they do 40

Read from the source code; the running server may list different tools.

Read 22 · Look things up without changing anything.

  • analyze_knowledge_base
  • auto_hook_query
  • detect_knowledge_gaps
  • find_path
  • get_auto_save_analytics
  • get_autonomous_config
  • get_config_value
  • get_memory
  • get_monitor_status
  • get_vault_info
  • get_vault_stats
  • list_available_tools
Show 10 more tools
  • list_config_all
  • list_memories
  • list_profiles_config
  • query_with_context
  • search_by_graph
  • search_memories
  • search_vault
  • suggest_links
  • suggest_tags
  • verify_last_save

Write 11 · Create, change, send or run things.

  • auto_hook_response
  • batch_update
  • bulk_create
  • create_backup_tool
  • create_memory
  • create_profile_config
  • import_backup_tool
  • import_document
  • relate_memories
  • set_config_value
  • update_memory

Delete 3 · Delete, drop, revoke or wipe things.

  • batch_delete
  • delete_memory
  • delete_profile_config

Not yet classified 4 · We could not tell from the name and description.

  • configure_autonomous_mode
  • export_vault_tool
  • get_save_stats
  • use_profile_config

Secrets and settings it asks for

Secrets: values you paste in that let it act as you.

  • ANTHROPIC_API_KEY
  • OPENAI_API_KEY

Other settings (environment variables or headers).

  • MEMOGRAPH_MODEL
  • MEMOGRAPH_PROVIDER
  • MEMOGRAPH_READONLY
  • MEMOGRAPH_VAULT
  • OLLAMA_HOST

Files it touches

  • Any path the agent passes to its tools

Tools (40)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

  • analyze_knowledge_base

    Perform comprehensive knowledge base analysis including gap detection, topic clustering, and learning path suggestions. Provides insights into the structure, completeness, and organization of the knowledge base with actionable recommendations.

  • auto_hook_query

    🔍 AUTO-SEARCH VAULT - Call at the START of a conversation to search for relevant context. ⚠️ NOTE: Usually it's better to just use 'search_vault' or 'query_with_context' directly. This tool: 1. Searches the vault for relevant memories 2. Optionally saves the query itself (usually disabled) 3. Returns context and sources Most users don't need this - use 'search_vault' instead.

  • auto_hook_response

    💾 SAVE CONVERSATION - Call this IMMEDIATELY after answering to save our exchange. ⚠️ IMPORTANT: This tool is NOT automatic - you must explicitly call it after each meaningful exchange. ✅ WHEN TO CALL: - After answering questions - After providing insights or advice - After planning or decision-making conversations - After any exchange worth remembering ❌ SKIP FOR: - Simple greetings ('hi', 'hello') - Acknowledgments ('ok', 'thanks') - Clarification questions 📝 USAGE: auto_hook_response( user_query='User's question here', ai_response='Your complete answer here' ) This builds conv…

  • batch_delete

    Delete multiple memories in one efficient operation. Much faster than calling delete_memory multiple times.

  • batch_update

    Update multiple memories in one efficient operation. Much faster than calling update_memory multiple times.

  • bulk_create

    Create multiple memories in one operation. More efficient than calling create_memory multiple times.

  • configure_autonomous_mode

    Configure autonomous hooks behavior (enable/disable auto-search, auto-save, etc.)

  • create_backup_tool

    Create a backup of the vault. Essential for disaster recovery and version control.

  • create_memory

    Create a new memory in the MemoGraph vault

  • create_profile_config

    Create a new configuration profile with custom settings

  • delete_memory

    Delete a memory from the vault by ID

  • delete_profile_config

    Delete a configuration profile

  • detect_knowledge_gaps

    Detect knowledge gaps in the vault including missing topics, weak coverage, isolated notes, and missing links. Provides actionable suggestions for improving the knowledge base structure and completeness.

  • export_vault_tool

    Export vault memories to JSON, CSV, or ZIP format. Useful for backups, data portability, and integration with external tools.

  • find_path

    Find the shortest path between two memories through the knowledge graph. Shows how two ideas connect through intermediate memories.

Show 25 more tools
  • get_auto_save_analytics

    📊 AUTO-SAVE ANALYTICS - Get comprehensive analytics on the hybrid auto-save system. Shows detailed metrics: - Overall save rate and performance grade (A+ to F) - Layer 1 (explicit AI saves) vs Layer 2 (monitor) breakdown - Total conversations saved vs estimated missed - Monitor statistics (if enabled) - Actionable recommendations for improvement Use this to understand auto-save effectiveness and optimize configuration.

  • get_autonomous_config

    Get current autonomous hooks configuration and recommendations

  • get_config_value

    Get a configuration value from ~/.memograph/config.yaml

  • get_memory

    Get full content of a specific memory by ID

  • get_monitor_status

    🔍 MONITOR DIAGNOSTICS - Check if Layer 2 conversation monitor is running. Shows: - Whether monitor is enabled and active - Detection statistics - Background loop status - Recent activity Use this to diagnose why auto-save isn't working.

  • get_save_stats

    📊 SAVE STATISTICS - Get statistics about auto-save success rate. Shows: - How many saves have succeeded vs failed - Save rate percentage - Memory counts by type - Status interpretation (excellent/good/poor/critical) Helps you understand if auto-save is working effectively.

  • get_vault_info

    Get information about the currently configured MemoGraph vault (path, name, configuration). Use this when you need to know which vault is being used.

  • get_vault_stats

    Get statistics about the currently configured MemoGraph vault (memory counts, types, tags)

  • import_backup_tool

    Import memories from a backup file (JSON or ZIP). Useful for restoring from backups or migrating vaults.

  • import_document

    Import a document (TXT, PDF, DOCX) into the vault

  • list_available_tools

    List all available MCP tools with their descriptions and categories

  • list_config_all

    List all configuration settings from ~/.memograph/config.yaml

  • list_memories

    List memories with optional filters

  • list_profiles_config

    List all available configuration profiles

  • query_with_context

    Retrieve relevant context from the vault for a question. If LLM provider is configured, can optionally generate an answer. Otherwise, returns context for the client's LLM to use.

  • relate_memories

    Create a wikilink connection between two memories in the graph. This builds the knowledge graph by linking related ideas.

  • search_by_graph

    Traverse the knowledge graph from a memory to find connected memories within N hops. Reveals how ideas connect through wikilinks.

  • search_memories

    Search the persistent memory vault for relevant memories. Use this BEFORE answering any question that depends on past decisions, preferences, or established facts. Returns up to top_k memory titles + snippets.

  • search_vault

    Search memories in the MemoGraph vault using semantic or keyword search

  • set_config_value

    Set a configuration value in ~/.memograph/config.yaml

  • suggest_links

    Suggest wikilinks for a note using semantic similarity and graph analysis. Analyzes content, graph relationships, and context to recommend relevant connections to other notes. Includes bidirectional link suggestions (notes that should link to this one).

  • suggest_tags

    Suggest tags for a note using AI analysis. Analyzes content using frequency, semantic similarity, structure, and related notes to recommend relevant tags.

  • update_memory

    Update an existing memory's title, content, tags, or salience

  • use_profile_config

    Switch to a specific configuration profile

  • verify_last_save

    🔍 VERIFY AUTO-SAVE - Check if your last conversation was saved. Use this to: - Confirm auto-save is working - Debug save issues - Get peace of mind that exchanges are being preserved Returns details about the most recent save or troubleshooting tips if none found.

Signals

Stars
4
Forks
1
Open issues
1
Weekly downloads
21
Last commit
Jun 25, 2026 (4 months ago)
Last release
Jun 25, 2026 (4 months ago)
Last package publish
Jun 25, 2026 (4 months ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name
io.github.Indhar01/memograph
Version
0.4.0
Auth
API key
Packages
pypi memograph
License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026
Cite this page

Memograph MCP server. Agent Mart. https://agentmart.dev/connectors/indhar01-memograph (last updated 2026-10-24)