A run is one execution of an Agent Automation, from the moment a trigger fires to the moment the work finishes. Runs are where credits are spent and results happen, so Cando keeps a complete record: every run, and even every firing that didn’t become one.

​
From firing to run

When a trigger comes due, that’s a firing. A firing that finds the agent free becomes a run. A firing that finds the agent busy on something else doesn’t execute; it’s recorded as skipped, with a note naming what was holding the agent. So a gap in the record always has an explanation, and no work happens silently.

​
Run outcomes

Every run ends in one of a few states:
  • Succeeded: it finished
  • Failed: it didn’t, and the run’s chat shows what happened. A run also fails if it can’t finish before its next firing is due: a job on a five-minute interval gets five minutes
  • Cancelled: you stopped it
  • Skipped: it never started because the agent was busy

​
Every run produces a chat

Each run leaves a run chat: the full account of what your agent did, step by step, which you can read and reply to. Run chats don’t appear in your sidebar (a five-minute Agent Automation would bury your own conversations in a day); you reach them from Agent history.

​
Agent history

Agent Automations → Agent history is the record: every run of every Agent Automation on the agent, searchable, with when it fired, its outcome, and what it cost in credits. Open any row for that run’s chat. When you’re wondering what your agent has been doing, or why something didn’t happen, this is the page.