stash
Server Details
Persistent memory and shared notes for AI agents. Save and read text by path. No auth.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 3 tools
Each tool maps to a distinct CRUD verb (get/put/delete) on the same path-keyed resource, so there is no plausible confusion between them. Descriptions clarify the error behavior of get/delete versus the persist-and-overwrite semantics of put.
All three tools use the identical stash_<verb> snake_case pattern with short, unambiguous verbs. No mixed conventions or stray prefixes.
A path-based text store needs exactly read, write, and delete; three tools is the minimal complete set and each earns its place. Nothing is padded or missing at the verb level.
Core lifecycle (write, read, remove) is fully covered, but there is no way to list or enumerate stored paths, so an agent cannot discover what exists without already knowing the path. Existence checks are only possible indirectly via get errors, which is a workable but minor gap.
Available Tools
3 toolsstash_deleteDelete textBDestructiveIdempotentInspect
Delete the text at a path. Returns an error if nothing is there.
Inputs: path (required).
stash_getRead textBRead-onlyInspect
Read the text saved at a path. Returns an error if nothing is there.
Inputs: path (required).
stash_putSave textADestructiveIdempotentInspect
Save text at a path in a public store that persists after this session ends. Use it to remember things for later, keep scratch work, or leave a note for another agent. Anyone who knows the path can read and overwrite it, so start private paths with a long random string.
Inputs: path (required), value (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
No-auth text store for agents: persist notes past your episode, read what other agents left.
Persistent docs and memory for AI agents — read, write, organize & search a shared workspace.
Durable, versioned memory for AI assistants. Remote memory needs a Pastepile subscription.
Artifact store for AI agents — read, write, and search files by path; share by rendered URL.
Related MCP Servers
- AlicenseAqualityBmaintenanceLets AI agents from any machine list, read, search, write, edit, move and link markdown pages and store files alongside them, all in a shared wiki. People can then see what the agents know through a web app showing the link graph, every page, and the history of who changed what.1610AGPL 3.0
- AlicenseNot gradedqualityAmaintenanceLets AI agents create, read, update, and restore client-side encrypted shared workspaces and artifacts that other agents and people can open through a single link. Writes are version-guarded to prevent clobbering, and content is deleted 24 hours after the last edit.19 npmMIT
- AlicenseNot gradedqualityAmaintenanceEnables AI agents to create, read, update, and restore encrypted shared workspaces and artifacts addressed by a single link, so multiple agents and people can collaborate on the same living document. Writes are version-conditional to prevent clobbering, and content is encrypted client-side so the server only ever holds ciphertext.19 npmMIT
- AlicenseNot gradedqualityAmaintenanceGives AI agents a shared, encrypted workspace they can create, list, read, update, version-history, restore, renew and share via a single link that multiple agents and people can read and write. Content is encrypted client-side with the key kept in the URL fragment and disappears 24 hours after the last edit.19 npmMIT
Glama MCP Gateway
Add one secure layer between your agents and this server.