first-customer-finder
Pass
Audited by Gen Agent Trust Hub on Nov 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves ingesting and analyzing content from external sources like forums and social media, creating a potential vector for indirect prompt injection attacks.
- Ingestion points: Untrusted data enters the agent context via the fetching of URLs and searching of public discussions as described in
SKILL.md. - Boundary markers: The instructions do not define specific delimiters or isolation techniques for the ingested content.
- Capability inventory: The skill has the ability to fetch web content and execute a local script (
scripts/generate_report.py) which performs file reads and writes. - Sanitization: The
generate_report.pyscript appropriately useshtml.escapeand URL scheme validation to sanitize data before it is rendered into the final HTML report.- [COMMAND_EXECUTION]: The skill invokes a local Python script (scripts/generate_report.py) to transform the collected data into a standalone HTML report. - The script uses only Python standard libraries and performs safe file operations within the workspace.
- Input data is sanitized using
html.escapeandurlparseto prevent cross-site scripting in the generated output.
Audit Metadata