astrofy-init

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the astrofy CLI tool for project inspection (astrofy inspect), initialization (astrofy init), and status verification. These commands are local to the project environment and are used according to the skill's stated purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific metadata and configuration files, which could theoretically serve as an injection vector if the project source is compromised.
  • Ingestion points: Reads package.json, lockfile, astro.config.*, and internal .astrofy configuration files.
  • Boundary markers: No specific text delimiters are defined; however, the skill relies on strict adherence to technical documentation and version comparison.
  • Capability inventory: Performs local file reads/writes and executes the astrofy CLI tool; no unauthorized network or administrative actions are present.
  • Sanitization: Explicitly mandates the use of static data analysis instead of dynamic imports for configuration files to prevent side-effect execution. It also requires JSON schema validation (Ajv) for all configuration inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 12:27 AM