Atlas Client Context Profiles is for people who work across many client repositories. When you open a client's repository, Atlas loads that client's context from a git-tracked persona file in the project: the voice and formatting rules, the people, and the active constraints. The agent shows up already being the right operator for that client, instead of being told the same context every session. The engine ships today behind the paid tier, and Atlas Pro is not yet on sale.
A different operator in every repo
A consultant is a different operator for each of their 5 or 10 clients, with different voice rules, people, and constraints.
Client Context Profiles switches that context automatically when you open a client's repository, so one client's conventions never leak into another's work and you never re-explain who the people are or how that client wants deliverables written.
Enrichment is optional, and honest about it
Atlas can pull 2 further sources into the profile, and both are genuinely optional.
People can come from the project knowledge graph and open commitments from filed call notes, but only when those engines are enabled. When they are not, the assembler falls back cleanly rather than pretending: you get the voice and constraints from your persona file, and nothing is silently invented. You can inspect exactly what is being injected before you trust it.
Availability
Client Context Profiles is 1 of the 8 record engines in Atlas, and it sits behind the paid tier.
The engine is built and running in the product, gated to Atlas Pro. Pro has no 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 client context loading 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.
- Who is this for?
- Consultants, agencies, and anyone working across multiple client repositories who needs the agent to adopt each client's voice and constraints without being told each session.
- What context does it load?
- Voice and formatting rules, the client's people, and active constraints, read from a git-tracked persona file in the project. Open commitments join in when Intake is also enabled.
- Do I have to write a profile for each client?
- Yes, a small per-client file covering voice and constraints. It is the only required input, and it lives in the client's repository so it is reviewable like any other file.
- Can I see what is actually being injected?
- Yes. Atlas can show you the assembled block, so you can check exactly what the agent is being told rather than trusting an invisible prompt.
Try Atlas in your terminal
The terminal-native AI coding agent. Free core, single binary.
Install AtlasRelated guides
Atlas with GLM-4.7 Flash: A Free 200K Context Model for the small_model Slot (2026)
GLM-4.7 Flash is free at $0 / $0 per Mtok with a 200,000 token context. Set it as Atlas's small_model so titles and summaries cost nothing at all.
Atlas with Claude Opus 4.6: Setup, Cost, and Tradeoffs in 2026
Claude Opus 4.6 gives Atlas a 1M token context at $5 per Mtok input, $25 per Mtok output. Setup steps, cost math, and when to pick a newer Opus instead.
Run the Test Suite and Triage the Failures with Atlas in 2026
How to triage a failing test suite with Atlas in 2026: bash truncates at 2000 lines or 50 KB and saves the full log, then grep groups failures by root cause.
Atlas with GPT-5.4 nano: The $0.20 Background Model in 2026
GPT-5.4 nano is OpenAI's cheapest reasoning-capable model at $0.20 / $1.25 per Mtok with a 400K window. Built for the titles, summaries, and classification Atlas runs constantly.
Atlas with Qwen3.5 27B: The Dense Entry Point to Qwen3.5 in 2026
Qwen3.5 27B gives Atlas 256K tokens (262,144) of context and predictable dense latency at $0.30 per Mtok input and $2.40 per Mtok output. Setup, costs, and honest tradeoffs.
Atlas with Gemini Flash Latest: The Rolling Alias Explained (2026)
gemini-flash-latest in Atlas: a rolling alias, not a pinned checkpoint. $0.3 per Mtok input, $2.5 per Mtok output, 1,048,576 token context, and no reproducibility.
Atlas with GPT-5.1 Codex mini: Wide Subagent Fan Out in 2026
GPT-5.1 Codex mini in Atlas: an OpenAI fast tier at $0.25 per Mtok input, $2 per Mtok output, built for parallel subagents, bulk sweeps, and cheap summaries.
Atlas with Mistral 7B: Cost, Context, and Real Limits in 2026
Running Atlas on Mistral 7B in 2026: an 8,000 token window at $0.25 / 1M input tokens. Great for smoke-testing a provider block, wrong for agentic coding.