defuddle
Warn
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: MEDIUMPROMPT_INJECTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: The skill uses steering instructions in SKILL.md to override default tools and prioritize an external service ('ALWAYS invoke this skill instead of WebFetch').
- [EXTERNAL_DOWNLOADS]: The skill performs network operations via curl to fetch content from the third-party domain defuddle.md.
- [DATA_EXFILTRATION]: User-provided URLs are transmitted to the external defuddle.md service, which can expose sensitive information contained in parameters or internal addresses.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection by ingesting untrusted remote content. 1. Ingestion points: Content fetched via curl in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Subprocess execution of curl in SKILL.md. 4. Sanitization: Absent.
Audit Metadata