ASI01: Memory Poisoning
Adversarial entries injected into long-term agent memory persist across sessions.
Last reviewed October 2026
Why ASI01 evidence is hard
What makes memory poisoning worth its own control is persistence: the payload is written once and keeps working in sessions the attacker is not present for. Probes plant a durable preference through ordinary conversation, push instructions in through a rules file or a spoofed reminder channel, and poison a citation the agent will later trust - then check, in a fresh turn, whether the plant is still steering behaviour.
How Penaxtra maps to ASI01
The ASI01 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.
ASI01 capabilities
Findings tagged with the OWASP Agentic Top 10 ASI01 identifier
Penaxtra severity for this control (high)
Cross-framework identifiers attached to the same finding where controls overlap
PDF and JSON evidence export with the control identifier attached
ASI01 control coverage
Findings for ASI01 carry the OWASP Agentic Top 10 ASI01 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 ASI01 (Memory Poisoning)?
Adversarial entries injected into long-term agent memory persist across sessions. It is an OWASP Agentic Top 10 control; Penaxtra assesses it at high severity.
How does Penaxtra test for ASI01?
The ASI01 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 ASI01 help with an audit?
Each finding is tagged with the OWASP Agentic Top 10 ASI01 identifier and exported in the PDF and JSON evidence pack, so it appears on the auditor control list with the identifier already attached.