competition-browser-persistence
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines a structured workflow for auditing browser persistence. All operations are focused on state inspection within a security research context, with no evidence of malicious behavior or obfuscation.\n- [PROMPT_INJECTION]: The skill involves the ingestion of untrusted data from browser environments (cookies, storage, service workers), which presents an indirect prompt injection surface. This is an inherent part of the skill's auditing function.\n
- Ingestion points:
SKILL.mdidentifies cookies, storage buckets, service worker registrations, and cache entries as targets for inspection.\n - Boundary markers: Absent; the skill does not specify the use of delimiters or 'ignore' instructions for content retrieved from browser state.\n
- Capability inventory: The skill utilizes the agent's browser interaction capabilities to record, mutate, and replay stateful sequences.\n
- Sanitization: Absent; there are no instructions provided for the sanitization or validation of the data extracted from browser storage.
Audit Metadata