defuddle

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install an external dependency from the npm registry using npm install -g defuddle (SKILL.md).
  • [COMMAND_EXECUTION]: The skill utilizes the defuddle CLI to execute parsing operations on user-provided URLs and perform file writes using the -o flag (SKILL.md).
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from external websites.
  • Ingestion points: Untrusted content is ingested from arbitrary web pages via the URL parameter in defuddle parse <url> (SKILL.md).
  • Boundary markers: The instructions do not specify any delimiters or safety markers to help the agent distinguish between the tool's output and potentially malicious instructions embedded within the web content.
  • Capability inventory: The agent has the capability to execute shell commands and write files, which could be exploited if malicious content is successfully injected into the context.
  • Sanitization: No sanitization or filtering of the extracted web content is performed before it is returned to the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 11:21 AM
Security Audit — agent-trust-hub — defuddle