content-extraction
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to crawl and extract content from arbitrary external websites. An attacker could place malicious instructions on a website (e.g., in hidden HTML comments or metadata) to influence the behavior of the agent during the extraction process.
- Ingestion points: Content from external URLs provided by the user for extraction (SKILL.md).
- Boundary markers: No specific delimiters or instructions to 'ignore embedded commands' are mentioned in the skill's extraction logic.
- Capability inventory: The skill possesses capabilities for file system write access (saving data and assets) and network access (crawling via browser automation).
- Sanitization: The skill description does not include steps for sanitizing or validating the content extracted from the DOM before saving it or using it in further tasks.
Audit Metadata