Memory that keeps its receipts
Extracted facts answered 84.4% of LongMemEval-S. Linking each fact back to the conversation it came from raised that to 89.6%, for 20% more storage.
Tagged
memory
← All postsTinyFish, now on CoreSpeed
Web agents, search, saved sign-ins and monitors from TinyFish, in every agent you use, through one CoreSpeed connection.
Xbrain: Give Your Agent a Sense of Your Taste
Turn your X bookmarks into searchable design context for your agent.
Continue a task in a different agent without re-explaining
What carries over when you move a task from Claude Code to Codex through CoreSpeed, memory, connections and policy, and what does not, the conversation itself.
How memory search works: hybrid retrieval, excerpts and result budgets
memory__search fuses exact and relaxed lexical matching, entity aliases and vector similarity into one ranking, then bounds each row and the result.
How to connect Codex to your apps and give it memory
Register CoreSpeed in Codex with codex mcp add, sign in, connect your apps, and add memory rules to AGENTS.md so Codex recalls decisions across sessions.
How to give your agent memory that survives sessions and clients
Two standing rules, a scope choice, and a habit of saving decisions give any agent memory that outlives the session and the client it ran in.
Linking memories: a small graph that keeps decisions connected
Memory links are directed edges with a kind and optional weight. They are idempotent to write, visible only when both ends are, and capped at 1,000 per memory.
Memory isolation with row-level security, not application code
Caller identity comes from auth, each memory carries its scope, and Postgres row-level security enforces the boundary. Nothing in the request body can cross it.
Pull requests for shared memory: proposals, archive and versions
A correction next to a stale shared memory does not stop the old one being retrieved. Archive it with a successor, or propose a fix for its author to merge.
Shared memory vs CLAUDE.md and AGENTS.md files
Instruction files hold the rules an agent follows, per repository or per machine and per client. Org-owned memory holds searchable facts that follow you across both. Use both.
What is agent memory? Shared, private, and owned by the org
Agent memory is durable, searchable state that outlives one conversation, client or agent. How it differs from a context window, and who owns it.
How we took LongMemEval from 80.0 to 94.2 without touching retrieval
LongMemEval-S climbed from 80.0 to 94.2 while retrieval recall at 10 stayed at 0.99. The gains came from evidence budget, reader generation, and instructions.
We retrieved the memory, then dropped the answer
A 200-question conflict run scored 0.800 at the parent-Memory boundary and 0.635 at the evidence boundary. The gap was in what reached the reader.