openclaw-executive-assistant-local

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). Outsider free text can enter the LLM context at runtime via the “Offline communications triage” workflow: the agent reviews exported .eml files from code-along/03-offline-communications-triage/eml/, whose body/content is authored by external correspondents (email senders) and is ingested as readable text for triage.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)