webpage-reader
Pass
Audited by Gen Agent Trust Hub on Apr 3, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill's workflow includes an instruction to install the 'defuddle' npm package globally ('npm install -g defuddle') if it is not found on the host system.
- [COMMAND_EXECUTION]: The skill uses the 'defuddle' command-line interface to parse URLs and supports writing the output to local files using the '-o' flag.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from the web. Ingestion points: Content extracted from user-supplied URLs (SKILL.md). Boundary markers: Not present; the skill lacks instructions to wrap untrusted content in protective delimiters. Capability inventory: Subprocess execution of CLI tools and file-writing capabilities (SKILL.md). Sanitization: Not present; the skill does not specify any validation or filtering of the fetched web content.
Audit Metadata