is-agentic

Pass

Audited by Gen Agent Trust Hub on Nov 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [External CLI Execution]: The skill uses npx is-agentic to run a technical audit tool. This involves downloading and executing a package from the npm registry at runtime, which is a standard method for utilizing development tools but introduces a dependency on the external package's integrity.
  • [Remote Service Communication]: The skill interacts with is-agentic.com via an API and MCP endpoint to fetch audit reports. This involves sending a user-specified domain or URL to an external service for analysis. This is expected behavior for a scanning tool, but users should be aware that target URLs are shared with the service.
  • [Indirect Prompt Injection Surface]: The skill instructs the agent to read recommendation and details fields from a remote JSON report and use them to "fix the issues" on a site. This creates a surface where external content from a remote server influences the agent's actions on local files.
  • Ingestion points: Audit reports fetched from https://is-agentic.com/api/v1/report or the CLI output.
  • Boundary markers: The instructions do not specify delimiters or warnings to treat the remote recommendations as untrusted data.
  • Capability inventory: The skill explicitly tasks the agent with modifying local code or configurations to improve scores based on the report content.
  • Sanitization: There is no mention of sanitizing or validating the recommendations before the agent applies them to the codebase.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 3, 2026, 11:48 AM