defuddle
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
defuddlenpm package via the commandnpm install -g defuddle. - [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the
defuddleCLI to parse content from URLs and potentially write output to the file system. - [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by ingesting and processing data from external, untrusted web pages.
- Ingestion points: External content is retrieved via the
defuddle parse <url>command described inSKILL.md. - Boundary markers: Absent; the instructions do not provide delimiters or directions for the agent to ignore instructions embedded in the fetched content.
- Capability inventory: The skill uses subprocess execution to run the CLI tool and includes file-writing capabilities.
- Sanitization: Absent; no sanitization or validation of the retrieved web data is specified before it is processed by the agent.
Audit Metadata