Skip to content

Common Issues#

Login issues#

Login fails with incorrect email or password#

  • Double-check that the email and password are entered correctly
  • Watch for capitalization and trailing spaces
  • If you have forgotten your password, use the reset link on the login page or contact your admin

No models visible after login#

Possible causes:

  • Your team has not assigned shared models
  • You have not added a personal model yet
  • Your account does not have sufficient permissions

Resolution:

  • Go to Model configuration to add a personal model
  • Or contact your admin to confirm whether shared models have been assigned

Tool issues#

AI is not calling tools#

Possible causes:

  • The current model does not support Tool Use
  • The relevant tool has not been enabled
  • Your description was not specific enough

Resolution:

  • Switch to a model that supports Tool Use
  • Confirm the relevant tool is configured correctly
  • Be explicit in your prompt, for example: "Use available threat intelligence to check this hash"

Tool call returns no response or an error#

Check the following:

  1. Is the tool enabled?
  2. Is the API key or token still valid?
  3. Is the service URL entered correctly?
  4. Does your account have permission to use this type of tool?

AWS tools return Access Denied#

This is usually not an AttackTrace issue — it means the AWS credentials you provided do not have sufficient permissions.

Suggestions:

  • Ask your admin to confirm that the appropriate read-only permissions have been granted
  • Check the required permission scope in the specific tool's documentation

Query results are truncated#

Common with Elasticsearch, Splunk, Confluence, and other tools that return large payloads.

Resolution:

  • Narrow the time range
  • Reduce the number of results
  • Add more specific filter conditions

Query times out#

Resolution:

  • Change "last 7 days" to "last 1 hour" or "last 24 hours"
  • Verify that small-scope queries work first
  • Gradually expand the search range

Model issues#

Local model unavailable#

If you are using Ollama or LM Studio, the local model service must be running and the model must be loaded correctly.

Model responses are very slow#

Possible causes:

  • The model itself is slow
  • Your prompt is very long
  • Multiple tools are being called at the same time

Try narrowing the scope of the question, or switch to a faster model.