Agentic Memory Management via MCP — Knowledge Graph for AI agents
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"agentic-memory-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# agentic-memory-mcp > Agentic Memory Management via MCP — Knowledge Graph for AI agents. [](https://pypi.org/project/agentic-memory-mcp/) [](https://github.com/alexandrerodenas/agentic-memory-mcp/actions) [](https://www.python.org/) A **Model Context Protocol (MCP)** server that gives AI agents persistent, structured, and queryable memory via a Knowledge Graph. Designed to be used standalone, embedded in any agentic framework, or served as an MCP tool server. ## Architecture ``` agentic_memory/ ├── core/ │ ├── graph.py # KnowledgeGraph — nodes, edges, JSON persistence │ ├── score.py # Score & ScoreStore — reputation / relevance scoring │ └── prune.py # AutoPruner — maintenance & auto-pruning ├── mcp/ │ └── server.py # MCP server (stdio transport) ├── cli/ │ └── main.py # Click CLI (memory-cli command) ├── api.py # MemorySkills — Python API (standalone use) └── hermes/ # Hermes Agent plugin integration ``` **Persistence:** JSON files (`memory_graph.json`, `memory_scores.json`) — no external database required. ## Installation ```bash # From PyPI pip install agentic-memory-mcp # Development git clone https://github.com/alexandrerodenas/agentic-memory-mcp.git cd agentic-memory-mcp uv sync ``` ## Quick Start ### CLI ```bash # Add a memory entry memory-cli --graph memory.json node add --id alice --label Person --content "Alice lives in Paris" # Search memory-cli retrieve --query "Paris" # Link two entries memory-cli --graph memory.json edge add --id e1 --source alice --target paris --label lives_in # Get stats memory-cli --graph memory.json stats # Prune old entries memory-cli --graph memory.json prune --max-nodes 500 --strategy oldest ``` ### MCP Server …
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.