mnemoverse-docs
Skill0 starsChanged 37 days agoMITView on GitHub
Guides a developer adding Mnemoverse long-term memory to their own app, agent or framework: picks the form (hosted MCP server, npm MCP server, REST API, Python SDK, Vercel AI SDK package) and points to the docs page. Use when the user asks how to integrate Mnemoverse or which package, endpoint or authentication to use, or says \"Mnemoverse SDK\", \"Mnemoverse API\" or \"add memory to my agent\".
- Reads credentials
- Installs packages
In this repository6 other files it gives its agents.
- agent-memory-disciplineSkill
- memory-catch-upSkill
- memory-onboardSkill
- memory-saveSkill
- memory-searchSkill
- setupSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
If the user wants memory in the assistant they are talking to right now, this is not an integration question.
- Distinctive: line 18
Every form reaches the same memory in the same account, so a fact written through one is recalled through the others.
- Distinctive: line 37
Every published package in one list: https://mnemoverse.com/docs/api/integrations
- Distinctive: line 38
When something fails: https://mnemoverse.com/docs/api/troubleshooting
- Distinctive: line 47
Keep the key out of code, chat and memory: an environment variable or the host's secret store.
- Distinctive: line 48
If the docs do not answer the question, say so and point to the closest page rather than guessing at behavior.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Mnemoverse docs for developers
This skill is a map of the Mnemoverse documentation for someone building with it. It needs no memory tools: it answers from the docs pages below.
Check first: building, or connecting this assistant?
If the user wants memory in the assistant they are talking to right now, this is not an integration question. Use the bundled setup skill if it is installed, otherwise https://mnemoverse.com/docs/api/install-by-host, and stop here.
Everything below is for a developer whose own code will read and write memory.
Pick the form
Every form reaches the same memory in the same account, so a fact written through one is recalled through the others. Pick by where the code runs and how it authenticates.
| The developer is building | Use | Docs |
| An MCP client that can open a browser sign-in with a loopback callback (a desktop or CLI app) | The hosted MCP server at https://mcp.mnemoverse.com/mcp, OAuth 2.1 with PKCE, no key to paste; a web client served from its own domain cannot register, so it uses the API key row below | https://mnemoverse.com/docs/api/remote-mcp-server#which-clients-can-sign-in-registration-allowlist |
| A server-side app that speaks MCP over HTTP but cannot do a browser sign-in | The same hosted server with a Mnemoverse API key sent as a bearer token | https://mnemoverse.com/docs/api/remote-mcp-server#with-an-api-key-instead-of-the-sign-in |
| An MCP client that starts local servers | The npm package @mnemoverse/mcp-memory-server, started with npx -y @mnemoverse/mcp-memory-server@latest and the key in MNEMOVERSE_API_KEY | https://mnemoverse.com/docs/api/mcp-server |
| Python code | pip install mnemoverse (Python 3.10 or newer), sync and async clients | https://mnemoverse.com/docs/api/python-sdk |
| A Vercel AI SDK agent | The @mnemoverse/ai-sdk package, or the MCP server through createMCPClient | https://mnemoverse.com/docs/api/vercel-ai-sdk |
| Any other language, plain HTTPS | The REST API at https://core.mnemoverse.com/api/v1 | https://mnemoverse.com/docs/api/reference |
| An agent framework (LangGraph, CrewAI and others) | The framework guide | https://mnemoverse.com/docs/api/use-cases/agent-frameworks |
The REST base and the MCP address are different surfaces: an MCP client pointed at the REST base cannot complete its sign-in. The first table on https://mnemoverse.com/docs/api/agent-setup puts the two addresses side by side.
First steps to give the developer
- A free account and API key at https://console.mnemoverse.com, no credit card. Every form in the table above takes this key except the browser sign-in to the hosted server.
- The first write, read and feedback call in a couple of minutes: https://mnemoverse.com/docs/api/getting-started
- Per-host install commands (Claude Code, Cursor, VS Code, Codex, Gemini CLI and others): https://mnemoverse.com/docs/api/install-by-host
- Every published package in one list: https://mnemoverse.com/docs/api/integrations
- When something fails: https://mnemoverse.com/docs/api/troubleshooting
What to say about how it works
Keep to what the docs state. Memory is written as self-contained statements, read back by relevance or newest first, and rated after use: tell it a recalled memory helped or misled, and it re-ranks what comes back next. Shared rooms (beta) let several accounts write to one pool: https://mnemoverse.com/docs/api/rooms. Prices and limits: https://mnemoverse.com/docs/api/pricing.
Rules
- Link the docs page instead of retyping long configs from memory; configs change, the page is kept current.
- Keep the key out of code, chat and memory: an environment variable or the host's secret store.
- If the docs do not answer the question, say so and point to the closest page rather than guessing at behavior.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.