web-article-translator
Pass
Audited by Gen Agent Trust Hub on Apr 5, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted content from external URLs provided by the user. Malicious instructions embedded in the source web page could attempt to influence or override the agent's translation behavior.
- Ingestion point: Output from the mcp__web_reader__webReader tool as described in SKILL.md.
- Boundary markers: The instructions lack delimiters or specific guardrail prompts to distinguish between article content and instructions.
- Capability inventory: The agent has the capability to write files to the local system.
- Sanitization: No sanitization or validation of the fetched web content is performed before the translation and file-saving steps.
- [COMMAND_EXECUTION]: The skill instructs the agent to perform local file system write operations to save translated articles to the current working directory.
Audit Metadata