Atlas Intake takes a call transcript and extracts the structured intent inside it: commitments, decisions, facts, people, and scope changes, each tagged with who said it and the line it came from. It then files each item where you keep that kind of note and remembers the choice, so it files the same way next time. The engine ships in Atlas today behind the paid tier, and Atlas Pro is not yet on sale.
From transcript to filed knowledge
A consultant's commitments and decisions live inside 1-hour call recordings and then rot, because turning them into filed notes is manual work.
Intake runs a single extraction pass over each transcript and pulls out the items that matter, each tagged with who said it and a link back to the line it came from. What was buried in an hour-long recording becomes a short list of commitments, decisions, and facts you can act on.
It learns your filing system
The first time Intake sees a kind of item for a client, it asks once where you want it filed, then remembers.
Commitments for one client might go to a checklist, decisions to a decision log, people to your knowledge base. Intake learns these routing rules per client and per item type, files new items automatically, and lets you correct any choice, which updates the rule. The agent learns your information architecture instead of being told it every time.
Drop a file in; a Fireflies pull is still to come
Intake works with any notetaker today through 1 path: dropping an exported transcript into the project.
The file-drop source is built and is what runs now. A Fireflies integration that pulls finished calls in automatically is designed but not yet implemented. The same extraction and filing applies either way, and the capability is opt-in per workspace.
Availability
Intake is 1 of the 8 record engines in Atlas, and it sits behind the paid tier rather than the free core.
The engine is built and running in the product. It is gated to Atlas Pro, and Pro does not have a checkout yet, so it is not something you can buy today. Shipped, the automated work log, is the one record engine available free right now.
Frequently asked questions
- Is Atlas transcript extraction available now?
- The engine is built and running in Atlas, but it sits behind the paid tier and Atlas Pro is not yet on sale, so you cannot purchase access today.
- Which notetakers does it support?
- Any notetaker, by dropping an exported transcript file into the project. An automatic Fireflies pull is designed but not yet built.
- What does it extract?
- Commitments, decisions, facts, people, scope changes, and action items, each linked back to the transcript line it came from.
- Do I have to tell it where to file everything?
- Only once per kind of item per client. After that it files the same way automatically, and you can change any rule later by correcting a filing.
- Which record engine can I use for free?
- Shipped, the automated work log. It reads your commits and sessions and is on by default in the free core.
Try Atlas in your terminal
The terminal-native AI coding agent. Free core, single binary.
Install AtlasRelated guides
Atlas with Claude Fable 5: Plan Mode Model Guide for 2026
Claude Fable 5 is Anthropic's premium June 2026 model at $10 / $50 per Mtok with a 1M window. Use it for one expensive Atlas planning pass, then drop back down.
Atlas with Phi-4 (local via Ollama): The 16K Context Tradeoff in 2026
Phi-4 (local via Ollama) drives Atlas at Free (self-hosted) pricing with a 16K tokens (16,384) window. Setup, the 14B reasoning case, and where it breaks.
Atlas with MiniMax-M2.7-highspeed: The Fast Lane in 2026
MiniMax-M2.7-highspeed gives Atlas priority serving on MiniMax's newest 230B MoE at $0.60 per Mtok input and $2.40 per Mtok output, with a 204,800 token context.
Add a Regression Test for a Bug Fix with Atlas in 2026
How to add a regression test with Atlas in 2026: red first, then green. bash records the exit code, write creates the failing test, and edit applies the fix.
Atlas with Qwen2.5 72B (local via Ollama): Air-Gapped Coding in 2026
Run Atlas fully offline on Qwen2.5 72B (local via Ollama) in 2026. Free (self-hosted), about 47 GB at Q4_K_M, and a 32,768 token local context cap.
Atlas with Kimi K2 Thinking: Open-Weights Reasoning at $0.60 per Mtok (2026)
Kimi K2 Thinking gives Atlas reasoning at $0.60 / $2.50 per Mtok on a 262,144 token window. Open weights, and it sustains the long tool-calling chains agents need.
Atlas with Qwen Plus: 1M Context for $0.40 per Mtok in 2026
Run Atlas on Qwen Plus in 2026. Alibaba's mid tier gives 1M tokens (1,000,000) of context with reasoning at $0.40 per Mtok input, $1.20 per Mtok output.
Atlas with Llama 3.1 70B (local via Ollama): A 43GB Private Agent in 2026
Llama 3.1 70B (local via Ollama) in Atlas for 2026: a 43GB pull for a 48GB GPU or 64GB Mac, Free (self-hosted), with 128,000 tokens of context on your own hardware.