Agent runs 

The inbox 

The module shows two lists:

• Awaiting your decision — runs paused for an approval or for input, each rendered as a card with the controls to resolve it.
• Recent runs — a read-only table of the most recently finished runs (configuration, status, created, finished, approval, cost when non-zero, and — on the runs you may open — a Timeline link to the run's detail page).

A run whose tool call failed still ends with the status completed: the model receives the error as the tool's result and answers, so the run itself completed (ADR-092, ADR-191). The Status cell then adds Tool calls failed: N, in the recent runs and in the background list, so a write that did not happen is not hidden behind completed. It counts the tool steps the run's timeline shows as failed — a denied approval and a refused call among them — and not a call that was cancelled. The timeline names each one. Dashboards and the run queue read the status and are not changed by it.

When the call you just approved fails, the message after the approval says so — Approval recorded, but the approved tool call failed — instead of Approval recorded; the run continued.

The Approval column states who granted a run's approval (ADR-173), so you can see it without opening each run. Four readings, and only Approved by the person who started the run is highlighted:

• Approved by the person who started the run — the decision and the request came from the same backend user. Allowed by default; whether it is refused is the Require a second approver setting (ADR-172).
• Approved by someone other than the person who started the run — two people were involved.
• Approved, but the record does not say who started the run — the run was started by a service account or another non-backend caller, so there is no backend user to compare the approver against. The approver is still recorded.
• Approved, but the record does not say by whom — the approving user was not recorded, so nothing can be said about the decision either way.

A run can pass several approval fences, and the column has one cell for it. It then shows the strongest single reading, not a summary: Approved by the person who started the run if any one fence was released by the initiator, otherwise Approved, but the record does not say by whom if any one approver went unrecorded. So a run whose second fence a colleague signed still reads as self-approved, and the run timeline is where the fences are shown one by one.

An empty cell says three different things and must not be read as "no approval was needed": the run passed no approval fence, or it was denied, or the approvals could not be loaded or decoded. The run's timeline tells them apart, on the runs you may open.

The same label appears on the approval row of the run timeline, next to the decidedBy uid. A denial never carries one: denying your own run is deliberate, not an anomaly.

The column shows the same label to everyone who can see the row. A user holding Approve suspended AI runs (agent_approve) lists other people's runs without being able to open them, and for those runs the label is all the column says — it never names a user.

If the store cannot be read, the page shows a warning box rather than a silently empty inbox — an empty list therefore means "nothing waiting", not "load failed".

The run timeline 

Timeline opens one run end to end, read-only. It shows the run's summary — correlation, status and why it ended, configuration, rounds, tokens, cost — and below it a single time-ordered list of everything the run produced:

• Step — a recorded loop step (a request, a model answer, a tool execution).
• Provider call — a telemetry row: which provider and model served it, how long it took, whether a cache hit or a fallback was involved, and the error class when it failed. These are joined to the run because since ADR-153 every provider call a run makes carries the run's uuid as its correlation id.
• Governance — a decision taken during the run: a tool the gate withheld, a guardrail block, an approval requirement, an injected-context refusal.

When a guardrail stopped the run, the summary says why under Stopped because, in the guardrail's own words. A background run has nobody waiting for that message, so this is where its author finds it.

You can only open your own runs; an administrator can open all of them. The approval grant lets you decide another user's waiting run, but not read its timeline — the Timeline link is therefore offered only on the rows you may open. Reaching a run you may not read directly by URL is indistinguishable from reaching one that does not exist: both send you back to the list.

A run started before this extension version has steps but no provider calls: its calls were traced individually and cannot be attributed retroactively.

Approving a tool call 

A run pauses for approval (status WAITING_FOR_APPROVAL) when the agent wants to call a tool that opts in to human approval. A guardrail verdict of approval_required follows a different path: it fails the run with a policy reason. It does not create a resumable approval card.

The approval card lists every tool call in the pending turn — the tool name and, in a collapsible Arguments block, the exact arguments the model proposed. A call whose tool is no longer registered is flagged.

One Approve or Deny covers the whole pending turn, not a single call. Denying refuses the pending calls and continues the model conversation with that refusal; the model may answer or propose another turn.

After a denial nothing is executed. The model is told that the approval was declined and by whom — the user who started the run, or another backend user — and answers from that (ADR-200).

When the parent page already holds a page with the proposed title, the preview of create_page_draft carries a warning line, and so does the preview of create_content_element_draft when the page already holds an element of that type with the proposed header, so a second copy is not approved by accident. The preview lines are in the language of the backend user the run acts as (ADR-213).

If the tool cannot determine whether another approver may read the record, the stored preview is withheld. A failed read check keeps that approval card and other waiting runs available, without exposing the stored preview or the failure details. The user who started the run may still read the preview captured under their own permissions.

Providing input 

A run pauses for input (status WAITING_FOR_INPUT) when the agent asks for typed data against a declared schema. The card renders a form with one field per schema property (text, number, integer or checkbox, with the field description shown). Submitting validates and coerces the values against the current schema; invalid input re-renders the form in place, keeping the text and checkbox states on that run's card and pointing at the error. Other waiting cards retain their own empty form. The run remains waiting for input.

While the run continues 

An approval, a denial or a submitted input continues the run straight away, in the same request: the remaining rounds run before the page answers. With JavaScript the card shows Continuing the run and counts the steps as they happen; when the run stops, the inbox reloads and shows the outcome as a message. The count is all the card shows: what the run did is on its detail page, for whoever may open it (ADR-276).

Each step sends a line, so a reverse proxy in front of TYPO3 does not cut the request off at its read timeout (fastcgi_read_timeout in nginx, read_timeout in Caddy) while the run is still working. A single step longer than that timeout is still cut off. When the connection ends before the run does, the card says that the run continues on the server and links to the list. Leaving the page does not stop the run either; the outcome is shown the next time the inbox is opened.

Without JavaScript the form waits for the whole continuation, as before, and behind a proxy a long one can end in a gateway timeout (HTTP 504) although the run finishes. The inbox then shows the run's state on the next load.

Optional run diagnostics 

The run persister keeps its documented fallback when a repository operation fails, even if the optional diagnostic logger throws. Refused writes and failed claims retain false; a failed begin() or findRun() retains null. Logging cannot make these operations succeed. Explicit validation exceptions still propagate. A configured logger is attempted once with the existing severity, message and context, without a fallback logging attempt. See ADR-225: Agent outcomes survive optional diagnostic failures.

The executor also retains its original outcome and exception when optional logging fails after cancellation, lease loss, a guardrail verdict, a run failure or a refused approval/input suspension. A failed run keeps its FAILED outcome and the exception that ended it. Ownership guards continue to protect a row reclaimed by another worker. Successful stored suspensions retain their resumable state.

Queued failure recovery also preserves its original failure, retry classification and ownership-refusal outcome when diagnostics fail. A failed logger cannot replace a not_retryable reason with retries_exhausted or interrupt dead-letter handling after a real dispatch failure. Existing write fences and guarded requeue/settlement continue to apply.

For background execution, batches and recovery, use the operational guides: