defuddle
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing the
defuddlepackage globally vianpm install -g defuddle. This is a standard external dependency authored by the skill vendor. - [COMMAND_EXECUTION]: The skill uses the
defuddlecommand-line interface to perform web parsing and metadata extraction tasks. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes and displays content retrieved from external URLs provided at runtime.
- Ingestion points: Untrusted data enters the agent's context when the
defuddle parse <url>command is executed on external websites (SKILL.md). - Boundary markers: Absent; the skill does not specify any delimiters or instructions to ignore embedded commands within the fetched content.
- Capability inventory: The skill uses a CLI tool to fetch and format web content into markdown, which is then processed by the agent.
- Sanitization: The content is cleaned of HTML boilerplate and navigation elements, but there is no logic to filter or sanitize the resulting markdown for malicious instructions.
Audit Metadata