paniolo-scan-remediate

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to fetch the @paniolo/cli package from the NPM registry during the scanning and remediation process.
  • [COMMAND_EXECUTION]: The skill executes shell commands via npx to run the @paniolo/cli tool with arguments for scanning and reporting in JSON format.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests JSON report data from an external CLI tool to drive automated file modifications, creating a potential attack surface if tool output is manipulated. Evidence chain: (1) Ingestion points: SKILL.md (Steps 1 and 4). (2) Boundary markers: The skill requires presenting a fix plan and obtaining user confirmation before applying changes. (3) Capability inventory: File system write access (editing repository files) and shell command execution. (4) Sanitization: None identified in instructions; relies on user oversight of the proposed fix plan.
  • [DYNAMIC_EXECUTION]: The use of npx --yes represents the dynamic loading and execution of remote code from the NPM registry at runtime.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 06:05 PM
Security Audit — agent-trust-hub — paniolo-scan-remediate