export-download-debugging
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides purely instructional content for debugging export failures. No malicious patterns such as obfuscation, exfiltration, or unauthorized command execution were found.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze browser console output and application logs to diagnose issues. While this involves ingesting untrusted data, the context is focused on debugging and does not pose a significant risk.
- Ingestion points: Browser console output and application logs (SKILL.md).
- Boundary markers: None explicitly mentioned for log data.
- Capability inventory: The skill suggests running local development commands like
pnpm typecheckandpnpm guardfor validation (SKILL.md). - Sanitization: Not applicable as the skill recommends manual or automated verification of the fix.
Audit Metadata