Its own notes
Reliable context, for every agent.
Structure it once, maintain it together, and retrieve it from any agent. What one person fixes, every agent learns.
Works with all your agents
- Claude
- ChatGPT
- Claude Code
- Any MCP client
- Cursor
- Grok
- Codex

Two agents. Two answers.
Pick a question, then switch Atlas on.
A query it wrote, counting paused accounts
Metric Customer churn (monthly)
Metric Customer churn (monthly)
main.Its own project notes
No source
- Pause the release pipeline.
- Redeploy the last green build.
- Post in #incidents.
Skill Deploy rollback
- Pause the release pipeline.
- Redeploy the last green build.
- Post in #incidents.
Skill Deploy rollback
A 2024 policy PDF, since replaced
A Slack thread from June
Page Refund approvals · Verified
Page Refund approvals · Verified
Kept by people. Read by agents.
Four kinds of context, each kept once.
Pages and skills have owners. Metric SQL and table notes change only through review.
Reads your warehouse, read-onlyBigQuerySnowflake
Month-end close
Properties
1. Lock the ledger
- Lock postings for the closing month by the second business day.
- Reconcile payouts against Payout schedule.
Refund approvals
Approval limits
Support can refund up to €1,000 without approval.
Refunds over €1,000 need a finance lead's approval in #refunds.
Card refunds
Card refunds reach the customer in 5 to 10 business days.
Tables
The schema and guidance agents receive for each warehouse table.
Core
payouts
One row per payout to a merchant.
Columns
- payout_id
- STRING
- Stable payout ID
- merchant_id
- STRING
- Merchant account
- settled_at
- TIMESTAMP
- When funds settled
- amount_eur
- NUMERIC
- Payout amount in euros
When to use
Use for payout timing and totals. Join merchants through merchant_id.
Customer churn (monthly)
SQL exampleSQL
WITH starting AS (
SELECT account_id
FROM analytics.core.accounts_monthly
WHERE month = @month
AND status = 'paying'
),
churned AS (
SELECT DISTINCT account_id
FROM analytics.product.events
WHERE event_name = 'subscription_cancelled'
)
SELECT COUNT(churned.account_id)
/ COUNT(starting.account_id) AS monthly_churn_rate
FROM starting
LEFT JOIN churned USING (account_id)Tables in this query
accounts_monthlyevents
Agents see only what their person can open.
Agents propose. People approve.
Governance built in. An agent's edit arrives as a change request. Approve each item, and all of them apply together.
Skills and pages keep their saved history. Open an earlier revision, compare it, and restore it as a new revision. Deleted pages can be restored from the trash.
Try it: approve both items.
KestrelChange requests
Define churn once for finance reporting
Finance reported August churn as 3.1% and 4.6%. One query counted paused accounts as churned.
LocationKnowledge / Workspace / Finance / Churn reporting
Open current version ↗Show full content
Ready to apply to the current page
How we report churnReport monthly churn as
cancelled subscriptions divided by all subscriptions.the Customer churn (monthly) metric. Paused accounts are not churned.Board reporting uses the same number.
- NameCustomer churn (monthly)DescriptionShare of accounts paying at the start of a month that cancelled during it.SQL definitionRead-only query example
+ WITH starting AS ( + SELECT account_id + FROM analytics.core.accounts_monthly + WHERE month = @month + AND status = 'paying' + ), + churned AS ( + SELECT DISTINCT account_id + FROM analytics.product.events + WHERE event_name = 'subscription_cancelled' + ) + SELECT COUNT(churned.account_id) + / COUNT(starting.account_id) AS monthly_churn_rate + FROM starting + LEFT JOIN churned USING (account_id)
Requested via Month-end close, August · 1 source
Every agent, one copy.
Each person connects their own agents over MCP. Every agent reads Atlas with that person's access.
- PNPriya NairFinance
- Claude
- ChatGPT
- TRTomás ReyesEngineering
- Claude Code
- Cursor
- MCMei ChenSupport
- Claude
- Slack bot
- JWJonas WeberData
- Codex
- ChatGPT
- ACAna CostaOperations
- Any MCP client
app.atlascontext.ai/mcp Each line is one person's own sign-in, with its own grant and never their browser session. Search checks access before it ranks, and every call checks it again.
See what gets read.
Analytics shows which pages are read, and whether agents or people read them.
Start with one team.
Bring one team's context. Book a demo, and we'll talk through your use cases together.
Invite-only pilot, hosted by Ensemble Labs in Paris.
Two ways in: Company Brain and Agent Memory.



