argmem-mcp
OfficialClick on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@argmem-mcpadd an observation that the API returned 500 to support the claim that the service is down"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
argmem-mcp
Official MCP server launcher for ArgMem — agent memory with receipts.
ArgMem is argumentation memory for LLM agents: observations (grounded facts with provenance) support claims; objections contest them; sustaining arguments answer objections; presumptions carry a TTL and quietly lapse unless re-confirmed. Belief states (HELD / CONTESTED / LAPSED / RETRACTED) are computed from the argument graph on every write — never asserted — and the whole store is an append-only JSONL audit log that replays to a byte-identical state on restart.
Install
npm install -g argmem-mcpRelated MCP server: aafp-commons
Use with Claude Code
claude mcp add argmem -- argmem-mcpor with any MCP client that runs a stdio command.
The engine ships separately
This package is the open launcher. The ArgMem engine is proprietary and is not distributed through npm:
Get access at https://argmem.com (free tier available).
Save the engine zipapp to
~/.argmem/engine.pyz(or pointARGMEM_ENGINEat it).Run
argmem-mcp.
The launcher finds a local Python 3 (python3 or python), starts the
engine in MCP stdio mode, and gets out of the way.
License
MIT for this launcher (see LICENSE). The ArgMem engine is licensed separately under Argmem's terms — https://argmem.com.
© 2026 Argmem · U.S. Provisional Patent Application No. 64/177,143
This server cannot be deployed
Maintenance
Related MCP Connectors
- memnodeOAuthdev.memnode
Persistent, inspectable memory for AI agents with lineage, correction, and a hosted MCP endpoint.
Governed personal world model and memory for your AI agent. Pair once, connect over MCP.
Persistent memory for AI agents across Claude, ChatGPT and any MCP client.
Private-by-default, local-first memory/context/task orchestrator for MCP apps and agents.
Related MCP Servers
- AlicenseAqualityCmaintenanceProvides a local-first, provenance-aware memory layer that enables MCP-capable AIs to store, recall, validate, and reason over facts with contradiction detection, trust weighting, deduplication, and encryption, supporting offline private operation without GPUs or API keys.9Apache 2.0
- AlicenseBqualityCmaintenanceEnables software agents to maintain a signed, evidence-aware local ledger of claims, with policy-gated proposals, conflict queries, and constitution-pinned access through an MCP stdio server.73Apache 2.0
- AlicenseNot gradedqualityAmaintenanceProvides a local stdio MCP server that gives AI coding agents a typed, temporal, evidence-backed memory on embedded SQLite, letting them record decisions and constraints, ask why something happened, navigate memory by time, trace proof paths between refs, and export or supersede stored state without rewriting it. It also exposes read-only ChronoLoom view tools so a person and their agent can steer the same multidimensional memory view by semantic intent rather than raw coordinates.Apache 2.0
- AlicenseNot gradedqualityCmaintenanceEnables any MCP client (zcode, Claude Code, Cursor) to give its model bi-temporal long-term memory, auditable evidence-cited retrieval, LLM-assisted document extraction, and deterministic A* skill planning with precondition/effect verification. All four capabilities share one portable SQLite store and can be mounted independently on demand.2MIT