Agent Did Not Run or Produce Output When Expected in Eve

Last updated: March 10, 2026

Overview

This article helps you resolve situations where an Auto Agent did not run or produce output on a matter when you expected it to. This typically occurs when trigger conditions are not satisfied, the matter has not been evaluated yet, cooldown periods are active, run limits have been reached, or the matter is in a state that prevents evaluation. This affects automatic agent execution and requires checking the Auto Agent Log to identify the specific cause.

Before You Start — Diagnosis

Confirm the following before attempting any fixes:

  • The agent is enabled and active (not paused or disabled in the agent settings)

  • The matter is open (agents do not evaluate closed matters)

  • You have checked the Auto Agent Log for this specific agent and matter to see the evaluation status

  • You understand what the agent's trigger conditions are (schedule type and 'Additional Condition' condition if configured)

  • At least one hour has passed since the agent was created or the triggering condition occurred (agents evaluate on the hour)

Fixes

Possible Cause: Trigger Condition Not Met

Cause: The agent evaluated the matter but determined the trigger condition was not satisfied. This is the most common reason an agent does not produce output. The agent checked the matter on schedule but found that the specific 'Additional Condition' condition you configured (such as "new medical records uploaded" or "new documents in Discovery folder") was not true for this matter at the time of evaluation.

Steps:

  1. Navigate to Agents in the left sidebar

  2. Select the Auto Agent Log tab at the top of the page

  3. Click the card for the agent that did not run

  4. In the Activity Details section, filter to Condition not met

  5. Find the matter in the list and expand the row

  6. Read the "Why the agent did not produce output" explanation — this tells you exactly what Eve was looking for and why it didn't find it

  7. Based on the explanation, determine if:

- The condition logic needs to be adjusted (e.g., folder name is incorrect, document type filter is too narrow)

- The expected documents have not actually been uploaded yet

- The documents were uploaded before the agent was created (trigger conditions only fire on new activity after agent creation)

  1. If the condition needs adjustment, go to the agent settings, edit the trigger condition, test it using the trigger testing panel, and save the updated condition

Expected Outcome: After reviewing the explanation and making any necessary adjustments to the trigger condition, the agent will evaluate and run correctly on the next evaluation cycle when the condition is truly met. If no adjustment is needed and the condition legitimately was not met, this confirms the agent is working as designed.

Possible Cause: Matter Not Yet Checked

Cause: The agent has not evaluated this matter yet in the current evaluation cycle. This is expected behavior shortly after an agent is created or when a matter first becomes eligible for evaluation. Auto Agents evaluate matters on a rolling hourly basis and do not retroactively process every matter in the firm immediately upon creation.

Steps:

  1. Navigate to Agents in the left sidebar

  2. Select the Auto Agent Log tab

  3. Click the card for the agent in question

  4. Filter to Not yet checked

  5. Confirm the matter appears in this list

  6. If the agent was just created or activated, wait for the next hourly evaluation cycle (agents evaluate at the top of each hour)

  7. If it has been more than 24 hours and the matter still shows "Not yet checked," verify the matter has had recent activity within the lookback window (24 hours for On Matter Update agents, 30 days for Monthly agents)

  8. To run the agent immediately without waiting, open the matter and run the agent manually from the agent panel — all Auto Agents can be run manually at any time

Expected Outcome: The matter will move out of "Not yet checked" status once the agent completes its next evaluation cycle. If the trigger conditions are met at that time, output will be generated. If you ran the agent manually, output appears immediately in the matter's Activity view.

Possible Cause: 24-Hour Cooldown Active

Cause: The agent already evaluated this matter within the past 24 hours. Eve includes a built-in safeguard that prevents any agent from re-evaluating the same matter more than once per 24-hour period. This prevents redundant runs and ensures agents do not repeatedly process the same matter unnecessarily.

Steps:

  1. Navigate to Agents in the left sidebar

  2. Select the Auto Agent Log tab

  3. Click the card for the agent

  4. In the Activity Details section, look at the Last Checked column for the matter in question

  5. Confirm the timestamp shows the matter was evaluated within the last 24 hours

  6. If you need output immediately and cannot wait for the cooldown to expire, open the matter and run the agent manually — the cooldown only applies to automatic triggers, not manual runs

  7. If the cooldown period has expired and the agent still has not run, check the other possible causes in this article (trigger condition not met, run limits reached, matter activity outside lookback window)

Expected Outcome: After 24 hours from the last evaluation, the agent will be eligible to evaluate the matter again on the next hourly cycle. If the trigger conditions are met at that time, output will be generated. Manual runs bypass the cooldown entirely and produce output immediately.

Possible Cause: Run Limits Reached

Cause: The matter has reached its maximum number of triggered runs for this agent or across all agents combined. Eve enforces two types of run limits as safeguards: a per-agent limit (if configured) and a global per-matter limit (default: 10 total triggered runs across all agents). Once a limit is reached, the agent stops evaluating that matter automatically.

Steps:

  1. Navigate to Agents in the left sidebar

  2. Select the Auto Agent Log tab

  3. Click the card for the agent

  4. In the Activity Details section, find the matter and check the Outputs Generated count

  5. Compare this count against the agent's configured run limit (if set) and the global per-matter limit (default: 10)

  6. If the limit has been reached and you still need the agent to run on this matter, contact your Firm Admin to request a limit increase for the agent or for the matter

  7. To produce output immediately without affecting automatic triggers, open the matter and run the agent manually — manual runs do not count toward triggered run limits

Expected Outcome: If the limit is increased, the agent will resume automatic evaluation on the next cycle. If you ran manually, output is generated immediately and appears in the matter's Activity view.

Possible Cause: Matter Is Syncing or Closed

Cause: The matter is currently syncing with a connected DMS or CMS system, or the matter is closed. Agents do not evaluate matters during active sync operations to avoid processing incomplete data. Agents also automatically skip closed matters, as they only evaluate open matters.

Steps:

  1. Open the matter in Eve

  2. Check if a sync indicator is visible at the top of the matter page — if present, wait for the sync to complete

  3. Check the matter status — navigate to the matter details and confirm whether the matter is marked as open or closed

  4. If the matter is syncing, wait for the sync to finish (typically completes within minutes) — the agent will automatically resume evaluation on the next hourly cycle after sync completion

  5. If the matter is closed and you need the agent to run, reopen the matter — agents will resume evaluation once the matter is open again

  6. To generate output immediately without waiting, run the agent manually once the sync completes or the matter is reopened

Expected Outcome: After the sync completes, the agent will evaluate the matter on the next hourly cycle. If the matter is reopened, automatic evaluation resumes immediately. Manual runs produce output as soon as the sync is complete or the matter is reopened.

Possible Cause: No Recent Activity Within Lookback Window

Cause: The matter has not had any document uploads or medical overview runs within the agent's lookback window. On Matter Update agents only evaluate matters with activity in the past 24 hours. Monthly agents only evaluate matters with activity in the past 30 days. Matters with no activity in these windows are automatically skipped to focus processing on active cases.

Steps:

  1. Navigate to Agents in the left sidebar

  2. Select the Auto Agent Log tab

  3. Click the card for the agent

  4. Check the agent's schedule type (On Matter Update or Monthly) to confirm the lookback window (24 hours or 30 days)

  5. Open the matter and check the Activity tab to see when the last document was uploaded or medical overview was run

  6. If the matter has been inactive beyond the lookback window, the agent will not evaluate it automatically — this is expected behavior

  7. If you need output on this matter despite the lack of recent activity, run the agent manually by opening the matter and selecting the agent from the agent panel

  8. If the matter should have recent activity but doesn't, verify that documents are uploading correctly and that any connected DMS/CMS sync is functioning

Expected Outcome: Once new documents are uploaded or a medical overview is run on the matter, the agent will include it in the next evaluation cycle (within one hour for On Matter Update agents, or on the next monthly schedule for Monthly agents). Manual runs produce output immediately regardless of recent activity.