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., "@.stashpad search my memories for the database decision we made"
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.
Memory for AI agents that you can read, edit and trust.
Stashpad keeps what an agent learns as plain Markdown files. Every memory has a source, dates and a trust level. Agents propose, a person accepts. Old facts are ended, never overwritten.
Plain files. Open them in any editor, diff them, keep them in git.
Safe by default. Agent and web writes wait in a review queue. Secrets are refused.
Works with any agent. Plain commands, an open MCP server and a plain AGENTS.md block. No model or company is required.
Optional encryption. stashpad encrypt seals every file with AES-256-GCM, snapshots included. Off by default.
No dependencies. Node 20 or newer is all it needs.
npm install -g stashpad # or run without installing: npx stashpad --help
stashpad save db-choice --kind decision --source-type user --source-ref "chat" --text "We keep data in SQLite."
stashpad search sqlite
stashpad setup # print the MCP settings for your agentAgents: read AGENT-GUIDE.md.
Start of a session, for any agent:
node bin/stashpad.mjs doctor # check the setup; says what to fix
node bin/stashpad.mjs brief # short summary, newest handover first
node bin/stashpad.mjs handover --text "Half done: auth. Next: tests." # note for the next session
node bin/stashpad.mjs sync-agents AGENTS.md # put the brief in a file agents already readRequires Node 20 or newer. No other dependencies.
npm test # run the tests
npm run build # build dist/
node bin/stashpad.mjs --helpMemories live in ./.stashpad/memories by default. Set STASHPAD_DIR or pass --dir.
Related MCP server: MCP Memory Server
stashpad save db-choice --kind decision --source-type user --source-ref "statement in chat" --text "We keep data in SQLite."
stashpad search "sqlite" --json
stashpad save db-choice-v2 --kind decision --source-type user --source-ref "later statement" --text "We keep data in Postgres." --supersedes db-choice
stashpad history db-choice-v2
stashpad search database --kind decision --as-of 2026-03-01 --explain # what was decided on that day, and why it matched
stashpad entities # names, tools and files the memories talk about
stashpad about Postgres # every memory about it, and what is mentioned with it
stashpad confirm db-choice-v2 # still true; doctor flags agent memories nobody confirmed for 180 days
stashpad handoff leave --text "Half done: auth. Next: tests." # a note the next session reads once
stashpad handoff claim # take it (a person's notes first); listing never takes one
stashpad feedback db-choice --signal stale # helpful, stale or wrong; wrong also asks for it to end
stashpad lint # what deserves a look; changes nothingThe command line, the MCP server and the review page can run at the same time: every write takes a short lock on the store.
Writes from agents, tools and the web wait in review:
stashpad review
stashpad accept <id>
stashpad reject <id> --reason "not verified"
stashpad audit{
"mcpServers": {
"stashpad": {
"command": "node",
"args": ["/path/to/stashpad/bin/stashpad.mjs", "mcp", "--dir", "/path/to/your/memories"]
}
}
}Eight tools: stashpad_search, stashpad_get, stashpad_recent, stashpad_about, stashpad_save, stashpad_forget, stashpad_handoff, stashpad_feedback.
stashpad eval # 40 notes, 33 queries (English and Dutch), safety checks; no language model is called. Paraphrased questions are measured too (about 2 of 3 found): word search misses some rewordings, and that is reported openly.
stashpad import notes.jsonl # queue memories from an export; they wait for review
stashpad export --out memories.jsonl # back up current memories as JSON Lines (--all adds ended ones)
stashpad save <id> --kind fact --source-type user --source-file notes.txt --text "..." # keeps a copy of the raw sourcestashpad review also shows when a waiting memory may conflict with a current one, and whether accepting it would end that one.
stashpad ui # opens http://127.0.0.1:4317/ to review, search and read the audit log
stashpad ui --open # the same, and opens the page in your browserOnly this computer can reach the page. It is a small local page with no outside resources. See the decision.
Imports come in as JSON Lines (id, kind, text, optional valid_from, source_ref) or as plain text, one fact per paragraph.
Plain files first. The files are the source of truth.
Nothing from an agent, tool or web page is stored until a person accepts it.
Secrets are refused at write time (keys, tokens, JWTs, Bearer tokens, passwords in connection strings).
Invisible characters are refused, and look-alike or split-up words cannot hide an instruction or a secret.
Instruction-like text (English and Dutch) is flagged for the reviewer.
A gatekeeper refuses agent notes that are too short or near-copies of an existing or waiting memory, and logs the refusal in the audit log. Notes a person writes are never blocked.
Limits: 4000 characters per memory, 200 writes waiting for review.
Recalled text is labelled as data, not instructions.
Third-party notices. Code is ported with npm run port, which refuses any license except MIT, Apache-2.0, BSD-2-Clause, BSD-3-Clause and ISC.
MIT. See LICENSE.
This server cannot be deployed
Universal memory for AI agents and tools. Save, organize and search context anywhere.
Persistent, inspectable memory for AI agents with lineage, correction, and a hosted MCP endpoint.
Persistent memory for AI agents. EU-hosted, privacy-first, hybrid recall, contradiction detection.
Persistent memory and knowledge graphs for AI agents. Hybrid search, context checkpoints, and more.
