An audit log is a time-ordered record of actions taken in a system, capturing who did what, to which item, and when, so events can be reviewed later.
What an audit log captures
Each entry records at least:
- Actor: the person, agent, or service.
- Action: created, edited, approved, published, deleted.
- Target: which item.
- Time: a precise timestamp.
- Context: inputs, reasons, or source references.
A useful audit log is append-only: entries are added but never edited or removed. Otherwise the log can be rewritten to hide a mistake. Stronger designs make tampering detectable, for example with a hash chain.
An example
A post goes out with a wrong date. The audit log shows the release bot drafted it at 10:02 from the v3.1 release notes, a reviewer edited the second line at 10:15, and the owner approved it at 10:16. The date error came from the release notes, not the agent.
Why it matters for AI-agent publishing
When several agents and people touch the same account, "who posted that?" needs a clear answer. It matters for fixing errors, for compliance, and for deciding whether an agent has earned more trust.
SocialHandoff keeps an append-only, hash-chained audit log that you can verify through the API. Entries carry citations, such as the GitHub release a draft came from. The trust ladder reads the same history of approvals. See the audit log guide and the compliance teams use case.