@conxt/mcp
Click 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., "@@conxt/mcp show me pending memory proposals"
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.
@conxt/mcp
Local-first MCP server for governed AI memory. Your agents propose memories; you approve them; only approved memories are ever recalled. Runs standalone with no account — optionally syncs to Conxt cloud for team memory.
Implements the open Governed Memory Record format.
Quickstart
Point your IDE at it — no install, no signup:
Cursor (~/.cursor/mcp.json)
{ "mcpServers": { "conxt": { "command": "npx", "args": ["-y", "@conxt/mcp"] } } }Claude Code
claude mcp add conxt -- npx -y @conxt/mcpClaude Desktop → Settings → Developer → Edit Config: same block as Cursor.
That's it. Memory lives in ~/.conxt/memory.json on your machine.
Related MCP server: Centralaizer
The tools
Tool | Who calls it | What it does |
| agent | Propose a memory (pending — doesn't govern yet) |
| agent | Recall approved memories for a query |
| agent/you | See what's awaiting approval |
| you | Approve a pending memory → it now governs |
| you | Decline or delete |
The rule that makes it governed: an agent can propose, but only a human can
approve — and recall never returns anything that isn't approved.
Human approval from the terminal
npx @conxt/mcp pending # what's waiting
npx @conxt/mcp approve <id> # approve it
npx @conxt/mcp reject <id>
npx @conxt/mcp list
npx @conxt/mcp path # where the store livesPrefer zero friction for solo use? Set CONXT_AUTO_APPROVE=1 and everything you
remember is approved on the spot.
Cloud mode (optional)
Local is always the source of truth. Set both env vars to also push approved records to your Conxt workspace for shared, governed team memory:
CONXT_API_KEY=... CONXT_API_URL=https://...A sync failure never affects your local memory.
Config
Env | Default | Purpose |
|
| Local store path |
| off | Approve on |
| unset | Optional cloud sync |
Develop
npm install && npm run build && node dist/index.jsLicense
MIT.
Publishing
CI (.github/workflows/ci.yml) builds and runs the governance smoke test on every
push. To release to npm:
On npmjs.com, confirm you own the
@conxtscope — or rename the package to unscopedconxt-mcpinpackage.json.Add an
NPM_TOKEN(Automation token) as a repository secret (Settings → Secrets and variables → Actions).Cut a release:
npm version patch # or minor / major — creates a v* tag git push --follow-tagsThe
publishworkflow builds, smoke-tests, and runsnpm publish --access public.
Or publish manually:
npm install && npm run build && npm login && npm publish --access publicAvailable Tools
6 toolsapproveA
Approve a pending memory (a human action). Once approved it governs and, if cloud sync is configured, is pushed to Conxt.
Inputs: id (required), approver.
forgetC
Delete a memory by id.
Inputs: id (required).
list_pendingA
List memories awaiting human approval, so they can be reviewed and approved or rejected.
Inputs: limit.
recallA
Recall approved memories relevant to a query. Only human-approved memories are returned — pending or rejected ones never surface.
Inputs: type, limit, query, scope.
rejectB
Reject a pending memory so it never governs.
Inputs: id (required).
rememberA
Propose a new memory. It is recorded as PENDING and does not govern until a human approves it (agents propose, humans approve). Set CONXT_AUTO_APPROVE=1 for frictionless solo use.
Inputs: type (required), content (required), scope, source.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
Each tool has a clearly distinct role in the memory lifecycle: remember proposes, recall retrieves approved, list_pending shows pending, approve/reject take opposite actions on pending, and forget deletes. No two tools appear to do the same thing, and the descriptions reinforce their unique purposes.
All tool names are lowercase verbs, but list_pending uses an underscore while the others are single words (remember, recall, approve, reject, forget). This is a minor deviation but the pattern is still predictable—most are simple imperatives. The inconsistency is small enough to not cause confusion.
With 6 tools, the server is well-scoped for a memory management system. It covers the essential operations without redundancy, and the count feels appropriate—not overly thin nor bloated.
The server covers create (remember), read (recall), and delete (forget) for approved memories, plus approval workflow tools (list_pending, approve, reject). However, there is no update/modify operation for existing memories, and no way to list all approved memories (recall only returns relevant ones). These gaps, particularly the missing update capability, mean the lifecycle is not fully complete.
Maintenance
Related MCP Connectors
Governed personal world model and memory for your AI agent. Pair once, connect over MCP.
Hosted memory for AI agents over MCP. One memory shared by all your agents.
Cloud-hosted MCP server for durable AI memory
One memory, every AI. A shared, user-owned markdown memory your AI clients read and write over MCP.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceA local-first MCP memory server providing persistent, searchable memory for AI agents, powered by SQLite.1 npm1Apache 2.0
- FlicenseNot gradedqualityCmaintenanceA local-first, privacy-preserving centralized memory hub for AI agents — MCP-compatible, zero cloud egress.3-
- AlicenseNot gradedqualityDmaintenanceA local-first memory server for AI tools that provides cross-tool, consent-gated, and auditable memory management via MCP.AGPL 3.0
- AlicenseNot gradedqualityCmaintenanceA local-first MCP server that gives AI clients a shared, user-owned memory and skills plane, enabling persistent memory across apps via the Model Context Protocol.MIT