illustration-ideas
Pass
Audited by Gen Agent Trust Hub on Apr 24, 2026
Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill retrieves data from external URLs provided by the user at runtime to analyze content for illustration opportunities.
- [COMMAND_EXECUTION]: The agent is instructed to execute specific commands using the
agent-browserCLI tool to interact with and capture information from web pages. - [EXTERNAL_DOWNLOADS]: The skill suggests that the user install the
agent-browserpackage via npm if it is not already available in the environment. - [PROMPT_INJECTION]: The skill processes untrusted web content, creating a risk of indirect prompt injection where malicious instructions on a webpage could influence agent behavior. Ingestion points: Web content from URLs in SKILL.md. Boundary markers: Absent; the skill does not use delimiters to isolate external content from instructions. Capability inventory: Interaction with browser tools to read text and HTML. Sanitization: Absent; the skill does not specify any validation or filtering of the retrieved web content.
Audit Metadata