Operating in real time
Watch the business run.
Your AI Cofounder helps move business work forward. This page shows durable activity when it exists, with a five-minute refresh window.
Last updated 2 minutes ago · platform aggregate · RSS · Reliability history
What your AI Cofounder is doing
- Businesses in flight2 plans
- Plans updated today0 plans
- Events emitted today0 events
- Tool calls in the last 24h0 calls
- Agents running cycles this hour0 cycles
Operating counters
Businesses built
Verified (independently checked by the platform against a real external system)2
0%Business briefs in the system
Plans in flight
Verified (independently checked by the platform against a real external system)2
0%Workspaces with at least one plan revision
Plans active (24h)
Verified (independently checked by the platform against a real external system)0
0%Plan revisions updated in the last 24 hours
Events (24h)
Mixed (blends independently verified rows with self-reported rows — only part of this count carries a receipt)0
trend unavailableBusiness-in-a-Box events emitted
Tool calls (24h)
Self-reported (not independently verified — no checker has run yet)0
trend unavailableAgent tool-call results observed
Cycles (1h)
Self-reported (not independently verified — no checker has run yet)0
trend unavailableHub cycle observations in the last hour (row volume, not a distinct-cycle count)
Recovery rate & cost per task
This view shows how often activity recovered after a failure and the recorded cost per task over the trailing 30 days. Availability and freshness stay visible when the read is incomplete.
Recovery rate
Mixed (blends independently verified rows with self-reported rows — only part of this count carries a receipt)14.3%
8 of 56 sessions that hit a real failure went on to succeed later in the same session.
Cost per task
Mixed (blends independently verified rows with self-reported rows — only part of this count carries a receipt)$0.00
$0.18 total across 60 metered tasks of 303 attempted.
Recovery rate and cost per task are computed from the durable record of every finished task in the trailing 30 days. Each entry is finalized from the same task-completion event the platform uses to classify failures and to meter billing — never anything the AI agent reports about itself. The system that writes these entries is fail-open: under heavy load it can drop the newest entry rather than delay a task, so a real finished task can occasionally be missing. That is why this is labelled mixed rather than verified. Recovery rate is the share of sessions that hit at least one real failure and later succeeded on a task in the same session — sessions with zero failures are excluded from both the recovered count and the total count, so an account with no failures never inflates the rate. Cost per task divides total recorded cost by only the finished tasks that carry a real recorded cost.
How we count
Every number on this page reads from durable storage. No ephemeral process, no per-request math. Refresh window: 5 minutes. A read that fails renders as Unavailable, never as a 0 — see the badge on each counter above.
Each counter also carries a provenance label. Verified means the number comes from an append-only, immutable database table — not something an agent narrated into a log. Self-reported means the opposite: the underlying table is documented, in the code that writes to it, as fire-and-forget observability rather than a validated receipt. Mixed means the count blends both kinds of row. We label a counter this way when applying the honest answer means it looks less impressive than "verified" — that is the correct outcome, not a bug.
- Businesses built Verified (independently checked by the platform against a real external system)
- The number of distinct business briefs across configured workspaces. Each brief is a structured description of the business a customer wants Armalo to help move forward.
- Plans in flight Verified (independently checked by the platform against a real external system)
- Workspaces that have at least one plan revision. A plan revision is a durable, versioned output of the planning stage.
- Plans active (24h) Verified (independently checked by the platform against a real external system)
- Plan revisions updated in the last 24 hours. Updates include new revisions, status changes, and edits.
- Events (24h) Mixed (blends independently verified rows with self-reported rows — only part of this count carries a receipt)
- Business-in-a-Box events emitted in the last 24 hours. Each event is a durable record of a state transition in the autonomous business.
- Tool calls (24h) Self-reported (not independently verified — no checker has run yet)
- Tool-call results observed in the last 24 hours. A tool call is one execution by the AI Cofounder against a connected tool, such as email, documents, or a query.
- Cycles (1h) Self-reported (not independently verified — no checker has run yet)
- Every hub_cycle_observations row recorded in the last hour — tool-call events plus raw agent narration and reasoning text. This is observation volume, not a count of distinct completed operating cycles.
Verified — derived from durable, append-only rows the system itself validated. Direct count of recorded business briefs. Each row is stored durably and cannot be changed after recording. Nothing is inferred or estimated.
Verified — derived from durable, append-only rows the system itself validated. Count of distinct workspaces with an active plan recorded. Plans use the same durable storage model as business briefs.
Verified — derived from durable, append-only rows the system itself validated. Count of plan updates recorded in the last 24 hours. This comes from plan timestamps, not activity volume.
Mixed — blends rows that require a durable receipt with rows that are internal bookkeeping only. Count of all recorded business events. Some events require an external record before they can be recorded. Others are internal progress updates. The total blends both kinds, so it is labelled mixed rather than verified.
Self-reported — derived from a fire-and-forget log the agent process writes as it runs, not an independently validated receipt. Count of tool-call results in the activity log. The AI Cofounder writes this log while it works, so some entries can be missing when a write fails. This counter does not replace the separate execution record.
Self-reported — derived from a fire-and-forget log the agent process writes as it runs, not an independently validated receipt. Count of operating activity entries in the last hour, including narration and reasoning. This measures activity volume, not completed work.
Ready to run one?
Your business can run itself.
Describe what you want to build. Your AI Cofounder builds, markets, and runs the business — handling sales, operations, customer success, and compliance so you can launch and grow without hiring.