Observed by Atlas — nobody has proved they control this domain. Is it yours?
Knowl
OBSERVEDPersistent, checkable memory for AI coding agents
- FRESHNESS
- 0.89
- DECAYS BETWEEN READS
- READINGS
- —
- VERIFIED / MEASURED
- VOUCHES
- 0
- SIGNED STATEMENTS
- THREADS
- 0
- FIELD NOTES
- CLICKS · 28D
- 2
- BOT-FILTERED
- FIRST SEEN
- AUG 28
- IN THE FLOW

Knowl is a tool that gives AI coding agents a shared, versioned memory so facts they store carry provenance, review status, and a successor when they change. It runs locally in a repository via MCP, keeping code and session history unuploaded until a fact is published to a shared workspace. It is aimed at teams whose agents currently rely on stale markdown notes or tribal knowledge.
Nothing measured yet. A reading is a measurement with its evidence attached — uptime checked, pricing re-read against what the page claims. Raise a checkable claim in a field note below and Atlas will measure it.
350 published facts a month · 1,000 in your first month · 5,000 stored facts · 1 workspace in the organisation, sharing the above · Unlimited members
2,500 published facts a month · 50,000 stored facts · 3 workspaces in the organisation, sharing the above · Unlimited members · 3 months of history
15,000 published facts a month · 300,000 stored facts · 10 workspaces in the organisation, sharing the above · Unlimited members · 1 year of history
Negotiated volume · Unlimited history · SSO and SCIM · Dedicated instance · Everything in Pro
Every crawl is diffed. Material changes are listed.
Nobody has vouched for Knowl yet.
A vouch says you use it. Your handle, how long, and an optional note.
I use this — vouchHave a question about Knowl? Ask the people using it.
No field notes yet. A field note is about use: what broke, what it replaced, what the docs don't say. Anything in one that makes a checkable claim gets measured by Atlas, which replies with the reading.
Outbound clicks are counted for the builder.
Claim it