LLM02: Sensitive Information Disclosure
Models leak PII, credentials, system prompts, or training-data fragments via responses or logs.
Last reviewed October 2026
Why LLM02 evidence is hard
A leak is one of the few things here that can be judged without interpretation: either the canary came back or it did not. Probes plant a credential, a national ID or an IBAN and look for it in the response, including through markdown image callbacks and links that exfiltrate when they render. Each canary is computed to pass its own checksum rather than invented, because an invented value never fires and the probe reports clean forever. The response body is never stored, only its digest.
How Penaxtra maps to LLM02
The LLM02 identifier is written onto the finding at the moment it is created, not mapped in afterwards, so the JSON and PDF exports come out already in your assessor's structure. Every finding carries the probe that was sent, a SHA-256 digest of the response it provoked, the tier that decided the verdict and its status history from first observation to close. Identity survives a re-scan: a control you closed and that comes back reopens under the same row with a reopen count, rather than arriving as a new issue.
LLM02 capabilities
Findings tagged with the OWASP LLM Top 10 LLM02 identifier
Penaxtra severity for this control (critical)
Cross-framework identifiers attached to the same finding where controls overlap
PDF and JSON evidence export with the control identifier attached
LLM02 control coverage
Findings for LLM02 carry the OWASP LLM Top 10 LLM02 identifier along with the corresponding control identifiers in the other frameworks Penaxtra maps, so one result is reflected across each mapped framework.
Frequently asked
What is LLM02 (Sensitive Information Disclosure)?
Models leak PII, credentials, system prompts, or training-data fragments via responses or logs. It is an OWASP LLM Top 10 control; Penaxtra assesses it at critical severity.
How does Penaxtra test for LLM02?
The LLM02 identifier is written onto the finding at the moment it is created, not mapped in afterwards, so the JSON and PDF exports come out already in your assessor's structure. Every finding carries the probe that was sent, a SHA-256 digest of the response it provoked, the tier that decided the verdict and its status history from first observation to close. Identity survives a re-scan: a control you closed and that comes back reopens under the same row with a reopen count, rather than arriving as a new issue.
Does a finding for LLM02 help with an audit?
Each finding is tagged with the OWASP LLM Top 10 LLM02 identifier and exported in the PDF and JSON evidence pack, so it appears on the auditor control list with the identifier already attached.