Wyrm

Most AI coding sessions start from zero. Wyrm gives the agent a memory that persists across them. It keeps your project's decisions, conventions, open work, and dead-ends in a structured database on your own machine, and hands them back to the model at the start of the next session. An agent connected to Wyrm recalls what was decided last week instead of re-deriving it, and can be stopped from repeating an approach that already failed.

It speaks the Model Context Protocol, so it drops into Claude, Cursor, Copilot, Windsurf, and Codex without glue code.

npm install -g wyrm-mcp # install wyrm-setup # wire it into your AI clients, then restart them

Then, from inside your client, ask it to callwyrm_capabilitiesto confirm the connection. The everyday loop is four steps the agent runs on its own once the habit sets in:

prime → load the project's truths, quests, and dead-ends at session start recall → retrieve what you know before re-deriving it check → before retrying an approach, ask if it already failed capture → store durable facts, lessons, and tasks as you go

Every number Wyrm publishes comes from a benchmark committed to the source, reproducible on your own data. The negative-learning firewall and the recall lift are the two that matter most, and both are covered below.

It remembers what failed, not just what worked

Most memory tools store successes. Wyrm also records dead-ends and blocks the repeat. You record a failed approach withwyrm_failure_record, and a laterwyrm_failure_checksurfaces it before the agent walks back into it. Across a session that stops re-litigating solved problems; across a fleet of agents, one worker's dead-end warns the rest, once.

Recall that finds things by meaning, not just keywords

wyrm_recallruns keyword search (FTS5) and semantic search over a vector index, fuses them, and reranks. It is hybrid by default, no configuration required. For higher accuracy you can opt into NVIDIA NIM retrieval (below).

By default nothing leaves your machine. The database is a single SQLite file at~/.wyrm/wyrm.db. When you do opt into a hosted embedding path, Wyrm reports exactly what left and where, in a determinism receipt and on its health endpoint. The privacy claim is one you can verify from the runtime, not just the docs.

Every memory is attributed to the agent and run that produced it, so a swarm of agents can share one accountable memory bus, with failures kept private to your account by default. A live event stream keeps devices in sync.

Wyrm can use NVIDIA NIM for embeddings and reranking when accuracy is worth a hosted call. On a retrieval benchmark committed in the repo, recall@1 moved from 33% on the local baseline to 47% with NIM embeddings and 52% with NIM reranking added. It is an explicit opt-in, off by default, and the egress is disclosed on every call.

export WYRM_VECTOR_PROVIDER=nim export WYRM_RERANK_PROVIDER=nim export NIM_API_KEY=nvapi-...

Ghost Protocol (Pvt) Ltd is a member of NVIDIA Inception.

Claude (Code, desktop, web) · Cursor · GitHub Copilot · Windsurf · Codex, and any MCP-capable client.

Node.js 22 or newer. Optional: a local Ollama withnomic-embed-textfor semantic recall without any hosted call.

Wyrm sends no telemetry, so the way we learn what is working is you telling us.

If Wyrm earned a place in your workflow, a star helps other people find it.

Persistent memory for AI assistants and coding agents across ChatGPT, Claude, Cursor, and other MCP-compatible tools.

One shared, versioned memory for your AI agents across Claude Code, Cursor, Kiro and Grok, with a source on every answer and full history.

Operational memory for AI agents that persists across sessions and tools.

MCP (Model Context Protocol) server that exposes 70 tools spanning cognitive memory, working memory, sessions, agents, the L0–L3 knowledge hierarchy, orchestration tasks, and training pipelines. The server speaks stdio and is consumed by Claude Desktop, Claude Code, OpenCodex, and any MCP-aware client.

MEMANTO is a memory agent. It remembers, recalls, and answers — so your agents can achieve long-term goals and avoid confusion.

A local-first, LLM-agnostic memory layer for AI assistants

Hosted MCP memory across Claude Code, Cursor, ChatGPT and any MCP client. Community-hosted instance of the MIT-licensed MemPalace engine. EU-hosted, GDPR-compliant. Free tier with 200 memories.

Shared memory layer for AI coding agents with dream pipeline distillation, session grounding, and multi-instance coherence.