invalidate

The invalidation layer for AI memory. Every fact gets a lease; new evidence ends it. Built on TypeSafe Jev.

How it uses Jev

invalidate uses Jev, a model that answers a yes/no question with a probability in about 150 ms. One check costs $0.00006. Checking 500 memories against one event takes under a second and costs half a cent. At that price, checking everything is the default.

Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README.

Keep exploring

Similar to invalidate

Browse all
Jev-Trades project preview
Applications
Trading bot with the all new TypeSafe AI's first system one model named as Jev
jev-lm (uhhfeef) project preview
Applications
A character-level language model built on Jev, a System One classifier
jselect project preview
Applications
Selects source-linked evidence within a token budget using Jev Noul relevance judgments and local diversity-aware selection.
jev_project_context project preview
Benchmarks & research
Evidence-first long-term experiment memory skill for AI coding agents, with optional Jev decision-model layers
jevmory project preview
Benchmarks & research
Coding-agent memory where every fact is a verbatim quote graded by TypeSafe Jev's calibrated confidence. Local-first, SQLite receipts, zero dependencies.
jev-compaction project preview
Demos & playgrounds
A context compactor that can only score, never write — so an agent's memory can't hold a fact the transcript never contained. Working demo, runs offline.
Jev-Mem project preview
Agent tooling
Jev-Mem: System-One Controlled Agentic Memory
typesafe-jev-workflow project preview
Async LangGraph workflow that gets a typed Jev Choice (invoice or general) and routes each inbound email to the matching handler.
jev-frontend-qa project preview
Browser & computer use
Evidence-driven frontend QA built on Jev Ultrafast and Browser Harness, with a synthetic todo demo.