Conversations, docs, and code context are noisy. Tenure turns useful signals into structured beliefs with scope, provenance, confidence, and a path for review before they become model-facing memory.
Tenure extracts decisions, preferences, facts, open questions, expertise signals, and style signals instead of treating entire conversations as memory.
Project, domain, team, and user scopes keep one task’s context from leaking into unrelated sessions.
In observation or curated modes, Tenure can extract and show what it learned without automatically giving it to the model.
Tenure separates signal capture from trust. Extraction creates candidates. Validation, scope enforcement, memory mode, and merge policy decide what happens next.
Tenure watches the relevant context: a conversation turn, onboarding input, imported documentation, or IDE workspace context.
The model proposes structured memory objects instead of free-form notes: new beliefs, updates, open questions, aliases, and style signals.
Tenure parses, repairs when possible, drops invalid output, and enforces the authoritative project scope in code.
Depending on your memory mode, extracted beliefs are ignored, saved as suggestions, merged automatically, or used only for insight.
A belief is not just an interesting sentence. It is a durable piece of state that can improve future model behavior when scoped and injected correctly.
RAG stores source text and retrieves passages later. Tenure extracts the durable claim, records where it came from, and gives future models the smallest useful belief instead of a transcript dump.
Later retrieval has to rediscover the useful fact from surrounding noise, stale turns, and unrelated context.
User: We switched validation to Zod... Assistant: Sure... User: Actually only at API boundaries...
The model receives the compact, scoped state it can use immediately, with provenance behind it.
{
type: "decision",
canonical_name: "zod_api_boundary_validation",
content: "Use Zod for validation at API boundaries.",
scope: ["project:billing-api"]
} The conversion point: Tenure can learn what it would remember before you allow that memory to affect model behavior.
Once work becomes structured belief state, teams can review it, merge it, supersede it, audit it, and decide exactly when it becomes model-facing memory.