ascendant

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes several local shell scripts (scripts/add-person.sh, scripts/check-transit.sh, scripts/ruling-planets.sh) to execute its astrological calculation engine. These scripts are internal wrappers for the skill's distributed TypeScript logic.
  • [EXTERNAL_DOWNLOADS]: The skill includes a setup script (scripts/setup.sh) that installs necessary Node.js dependencies, such as astro-ascendant and effect. These packages are official libraries associated with the vendor or reputable software organizations and are essential for the skill's stated functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes birth data and person records stored in the persons/ directory and provided via user attachments.
  • Ingestion points: Birth data input fields and files like input.md and MEMORY.md within the persons/ directory.
  • Boundary markers: The skill uses clear instructions to the agent to treat records as data, utilize markdown fences for summaries, and maintain the birth header as immutable.
  • Capability inventory: The skill performs filesystem writes to the persons/ directory and executes internal shell scripts for calculations.
  • Sanitization: Input data is parsed as structured fields (e.g., ISO 8601 moments, coordinates) before being used by the calculation tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 01:07 PM