Blog

Real-world use cases.

Concrete scenarios and workflows — what becomes possible when people and agents build on shared, proven knowledge instead of re-deriving it every time. Roughly weekly.

Latest

LLM-as-a-judge in production: what changes when the judge runs unattended

The published LLM-judge discipline — a calibration set, an expert to agree with, a dashboard someone reads — assumes a person nearby and a run that ends. knomit's judges run unattended in every maintenance pass, and their verdicts act: they merge and retract facts. Six things the nearby human used to provide have to be engineered into the judge itself.

The knomit team · 2026-08-28

Aug 15, 2026

Retraction is not deletion: what happens when an agent memory deletes a source

Agent-memory products ship add/update/delete as if memory were a CRUD store. But a memory that derives has downstream state resting on what it deletes, and one delete destroys both the record of what was believed and the means of finding what depended on it. What that costs, what Graphiti already fixes, and the two rules a store needs instead: every reference resolves at the commit it was written on, and a retraction may assert falsehood but never that history should leave the corpus.

knowledge
provenance
agents

Aug 8, 2026

Consensus by merge: a second road past "Don't Build Multi-Agents"

The multi-agent discourse converged on two safe roads — share full context between agents, or keep writes single-threaded — and for the artifacts it studied, mostly code, both roads are right. knomit's agents take a third: full writers running concurrently on different machines with no shared session context at all, not conflicting because each writes its own git branch and consensus is a merge. When can agents write in parallel without sharing context? When the artifact they write has a semantic merge operation.

multi-agent
agents
git

Jul 28, 2026

What an OKF bundle actually contains — and four things to do with it

One command turns a knomit knowledge base into 566 markdown files. The docs stop at git push; this is what you have afterwards — the anatomy of a bundle, and four ways to spend it, including Gemini and Claude both answering a design question straight off the files, with no knomit, no server and no MCP.

okf
publishing
interop

Jul 15, 2026

Does a knowledge base actually help a coding agent? An A/B experiment

The same task, with and without the knowledge base, eight runs per arm, judged blind. The assisted arm was always faster and cheaper — but one subtly flawed fact made it worse than no knowledge at all. Repairing that single fact flipped 30 of 32 blind comparisons and prevented a mistake models invent unprompted.

dogfooding
evaluation
agents

Jul 2, 2026

A guess is not a fact: a type system for knowledge

Store a model's speculation in the same shape as a verified observation and retrieval will grant them equal authority — that is how a guess gets laundered into a decision. knomit's answer is the move programming languages made decades ago: type every fact by its epistemic status and its origin, and enforce what may be derived from what. A guess can be stored, retrieved, even confirmed later — but it cannot pass as an observation.

knowledge
provenance
epistemics

Jul 2, 2026

Git is not storage: the REST moment for agent knowledge

Twenty-five years ago the industry tunneled RPC through HTTP POST until REST pointed out that the protocol already had the semantics. Agent memory is making the same mistake today — rebuilding provenance, history, identity, and sync inside vector stores, or auto-committing files into git while ignoring what its operations mean. knomit's bet: a commit is an assertion, a branch is an identity, a merge is consensus. Use git the way REST uses HTTP.

knowledge
git
provenance

Jun 28, 2026

When the code moves: a knowledge base that introspects

You shipped the feature and merged the PR — and somewhere in your knowledge base, a dozen facts just quietly went out of date. The fix isn't to write more carefully; it's to make the corpus introspect on itself. Here is a real, interactive pass where Claude Code swept knomit's own knowledge base fact-by-fact, checked every claim against the source instead of trusting the corpus, and corrected the drift — recording each correction instead of erasing it. This is the learning loop closing.

dogfooding
knowledge
provenance

Jun 27, 2026

Dogfooding: how knomit holds its own shape

knomit's own repo runs on knomit. We started where everyone starts — a fat CLAUDE.md and a per-machine MEMORY.md — and watched them rot and crowd out context. Here is what replaced them: typed, scoped knowledge surfaced at the moment of the decision, and one real fact that climbed from hunch to proven.

dogfooding
knowledge
provenance