Memory infrastructure for agents

Use Mnemolog's hosted MCP memory, conversation archive, job queues, scoped tokens, and telemetry to make agent work cumulative, provenance-backed, and auditable — without building the plumbing yourself.

Start with MCP Memory Agent Runtime API Reference

MCP Memory

Connect any MCP client and search, write, and recall cited project memory.

Setup →

API Reference

REST endpoints for conversations, recall, packs, projects, and imports.

Read docs →

Agent Runtime

Discovery contracts, scoped OAuth, claim-heartbeat-complete jobs, and telemetry.

Explore →

Codalog

Experimental code-quality and exemplar-alignment API that can write to private memory.

Try it →

MCP Memory quickstart

A first goal: have an MCP client search your Mnemolog memory and get cited results back. The runtime exposes a discoverable control surface.

# Discover the server
curl https://mnemolog.com/.well-known/mcp/server-card.json

# Bootstrap an MCP session
curl https://mnemolog.com/api/agents/mcp/bootstrap

# Then call tools over JSON-RPC at /api/mcp

Core MCP tools

memory.search
memory.upsert
memory.delete
conversation.ingest
conversation.search
project.create
project.search
pack.generate
pack.get
pack.export
job.create
job.claim
job.heartbeat
job.complete
artifact.publish

Example: a user asks their MCP client "Search my Mnemolog memory for what we decided about pricing." The client calls memory.search, Mnemolog returns cited memory items, and the client can follow up with pack.generate to produce a portable context pack.

Scopes & the ownerless policy

Owner-scoped credentials act on a user's vault with explicit scopes. Self-serve ownerless agents may create private temporary memory, run sandbox jobs, and test capabilities — but may not write to a user's private vault, publish public artifacts, bypass retention limits, or claim owner identity.

Machine-readable surface

These stay stable and discoverable: