If you want your AI assistant to remember things across sessions, the usual options — Mem0, Zep, Letta/MemGPT, Supermemory — are capable, but they all ask for an account, an API key, and usually an SDK before you can store a single memory. Corvid takes the opposite approach: one click gives you a private space and token, and a single MCP URL is the entire integration. This page is an honest comparison so you can pick the right one.
| Corvid | Mem0 / Zep / Letta | |
| Signup / account | None — one click | Required (email, dashboard, API key) |
| Integration | One MCP URL, or plain HTTP | SDK / client library, keys in env |
| MCP-native | Yes — remote streamable HTTP | Varies; usually SDK-first |
| Search | Ranked keyword (any term) | Vector / semantic |
| Price | Free | Free tier + paid plans |
Corvid uses keyword search today, not semantic/vector recall — so a query has to share words with the stored memory. There are no per-user sub-namespaces inside a space yet, and memories aren't encrypted at rest, so don't store secrets. If you need those, a heavier tool is the better fit. If you just want durable, no-friction memory your agent can call every session, Corvid is the fastest path.
curl -s -X POST https://corvid.perch-app.workers.dev/api/spaces
# -> {"id":"...","token":"cv_..."}
curl -s https://corvid.perch-app.workers.dev/api/memories -H "authorization: Bearer cv_..." \
-H content-type:application/json -d '{"text":"User ships with Cloudflare Workers"}'
curl -s "https://corvid.perch-app.workers.dev/api/memories?q=cloudflare" -H "authorization: Bearer cv_..."
Or set it up in Claude Desktop / Cursor via the MCP guide.