social-media-review
Pass
Audited by Gen Agent Trust Hub on Mar 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's primary functions, which include searching for public social media handles and providing step-by-step cleanup advice, are implemented securely. No hardcoded credentials, malicious remote code, or unauthorized data access patterns were found. The skill author's use of a configuration file (career-helper-preferences.md) is a legitimate implementation of user preferences.
- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection because it ingests and processes untrusted data from external social media platforms. 1. Ingestion points: Web search results and fetched profile content as specified in the research approach of references/social-scan-methodology.md. 2. Boundary markers: The instructions do not define explicit delimiters or 'ignore embedded instructions' warnings for the agent when processing fetched web data. 3. Capability inventory: The skill uses WebSearch, WebFetch, AskUserQuestion, and file writing capabilities (to save review and cleanup documents). 4. Sanitization: There is no evidence of sanitization or filtering of the external data before it is processed by the agent.
Audit Metadata