defuddle
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the 'defuddle' package globally via 'npm install -g defuddle'. This involves downloading and executing code from the public npm registry. While the skill identifies itself as part of the Galyarder Framework, the package is unpinned (no specific version) and retrieved from a non-whitelisted source.
- [INDIRECT_PROMPT_INJECTION]: The primary function of the skill is to parse web content from user-provided URLs. This presents a vulnerability surface for indirect prompt injection.
- Ingestion points: Web pages fetched via 'defuddle parse '.
- Boundary markers: No delimiters or specific instructions are provided to help the agent distinguish between the fetched content and its own instructions.
- Capability inventory: The skill has access to shell execution ('defuddle'), global package installation ('npm install -g'), and file system writes ('-o content.md').
- Sanitization: Although 'defuddle' is designed to remove 'clutter' (ads/navigation), it does not sanitize the text for malicious instructions meant to influence the LLM.
Audit Metadata