skills/paniolo-ai/scan/paniolo-scan/Gen Agent Trust Hub

paniolo-scan

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the @paniolo/cli package. This is a vendor-owned package belonging to the skill's author (paniolo-ai) and is fetched from the standard npm registry.
  • [COMMAND_EXECUTION]: Executes shell commands via npx to perform repository scans, capture diagnostic data in JSON format, and re-run scans to verify remediations.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes diagnostic data from external tool outputs (JSON reports) to guide the agent in performing automated file modifications within the repository.
  • Ingestion points: Findings report generated by @paniolo/cli scan (SKILL.md).
  • Boundary markers: Not explicitly defined for the ingested JSON output.
  • Capability inventory: File system writes (remediation edits) and shell command execution (scanning tools).
  • Sanitization: None mentioned for the tool findings before they are presented to the agent for remediation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 08:00 AM