kol-creator-discovery
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill performs network requests exclusively to the official vendor API (api.aisa.one) to retrieve creator contact information. It only transmits user-provided social media URLs and does not access sensitive local files.
- [COMMAND_EXECUTION]: The provided Python script uses standard libraries and argparse for input handling. It avoids dangerous execution patterns like eval() or shell command injection.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources. 1. Ingestion points: Social profile URLs and API responses from api.aisa.one. 2. Boundary markers: Data is confined to structured Markdown table output. 3. Capability inventory: Network requests via urllib and local file writing. 4. Sanitization: Employs md_escape to sanitize data for Markdown table integrity.
- [CREDENTIALS_UNSAFE]: The skill correctly handles sensitive credentials by requiring an AISA_API_KEY through environment variables and explicitly instructing the agent not to log or reveal the key.
Audit Metadata