Audit log

The audit log is your workspace's record of who did what, and when — teammates joining and leaving, API keys and webhooks being created or removed, board and roadmap privacy being flipped, moderation decisions, and plan changes. It's the page you open when someone asks "who changed this?", and the export you attach to a security questionnaire.

Find it under Settings → Audit log in the dashboard.

What gets recorded

Area	Recorded activity
Members	Invited, joined, removed, left, and role changes (with the old and new role)
API keys	Created, rotated, deleted — by key name, never the key itself
Webhooks	Endpoint created, updated, deleted — recorded by destination host, never the full URL
Boards	Created, deleted, and privacy switched between public and private
Roadmap	Privacy switched between public and internal — for the whole workspace or one product
Moderation	Each approve/reject decision on a held post
Identity	The customer-identity secret being generated or rotated, and the "reject unsigned calls" switch — never the secret itself
Plan	Plan changes, including the automatic move when a trial ends

Each entry records when it happened, who did it (the teammate's email, an API key's name, or "System" for automatic actions), what was affected, and — for actions taken by a person — the IP address the request came from. Open any row for the full detail, including before/after values where the action changed a setting.

Entries are written by us, never by hand, and they can't be edited or deleted from the dashboard.

How much we vouch for each entry

Open a row and you'll see an Attribution line. It says how the person named on that entry was established — not a detail most people need, but the thing that decides whether an entry is usable as evidence, so we state it plainly rather than implying every entry is equally certain.

Attribution	What it means
Verified	We established who acted ourselves — from their signed-in session, or by confirming they're a member of this workspace. This is the normal case for every entry.
Unverified	The action was reported to us with an actor we could not confirm against your workspace's membership. The entry is still kept, marked, and shown — never quietly dropped.

An "Unverified" entry is rare and worth asking us about. We record it rather than discarding it: hiding an anomaly from the log the anomaly belongs in would be the wrong trade.

How far back you can see

Your plan sets how much history the page shows you:

Plan	Audit log view window
Free	Not available
Startup	Last 30 days
Team	Last 1 year
Enterprise	Unlimited

We start recording your workspace's activity from day one, whatever plan you're on — the plan controls how far back you can look, not what we keep. Upgrade and the older entries that were already recorded become visible; they were never thrown away. When your plan is hiding older activity, the page tells you so directly.

Filtering and exporting

Narrow the list by event type, by who did it, or to a date range (the From/To fields are inclusive of both days). "Load more" walks further back through your window.

Export CSV downloads exactly what your current filters show — same rows, same window — with one row per entry and the full before/after detail in a metadata column. Values are escaped so a spreadsheet can't execute anything in them on import. Very large exports are refused with a message asking you to narrow the date range; narrow it and try again rather than retrying the same request.

If an export can't be built (a temporary problem on our side), you'll get an error rather than a partial file — an export that quietly reports "no activity" would be worse than no export at all.

The page applies the same rule to itself: if we can't load your activity, it says so instead of showing an empty table. An empty audit log and an audit log we failed to read look identical, and only one of them means nothing happened.

The audit log is available from the Startup plan up.