Outside content treated as data, not instructions
Pages, emails and files the agent reads cannot direct its actions.
Layer L2 Harness · Owner Agent framework vendor · Family Input and output safeguards
–Held
–Failed
–Got around
–Missing
By risk
| Risk | Held | Failed | Got around | Missing | Not mentioned |
|---|---|---|---|---|---|
| Hidden instructions hijack the agent | – | – | – | – | 5 |
Why it did not hold
No incident yet records this control not holding.
By quarter
| Period | Held | Failed | Got around | Missing |
|---|---|---|---|---|
| 2025-Q3 | – | – | – | – |
| 2025-Q4 | – | – | – | – |
| 2026-Q1 | – | – | – | – |
| 2026-Q2 | – | – | – | – |
| 2026-Q3 | – | – | – | – |
| 2026-Q4 | – | – | – | – |
Added in reported fixes, and what followed
| Incident | Developer | Date | Later incidents where it did not hold |
|---|---|---|---|
| LR-2026-0047 Varonis phishing simulation drove an OpenClaw email agent to forward credentials and a customer export to an external address | 2026-06-09 | None on record |
Frameworks
| Framework | Item | Text |
|---|---|---|
| MITRE ATLAS | AML.M0030 | Quoted from the stored text |
| OWASP LLM | LLM01 | Quoted from the stored text |
| OWASP Agentic | ASI01 | To verify when the text is stored |