astrofy-sitespec-implement

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the local command astrofy sitespec list --json to retrieve task details and verify that manual file edits conform to the expected schema. This is a legitimate use of a project-specific utility.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from SITEPEC.md files, including task descriptions and checklist items, which are then used to determine implementation steps and delegate tasks to other skills.\n
  • Ingestion points: SITEPEC.md (Description, Checklist, and Notes sections) and the output of the astrofy CLI tool.\n
  • Boundary markers: Absent. The instructions do not provide delimiters or warnings to isolate external task content from the agent's core instructions.\n
  • Capability inventory: File system write access for updating specification files, shell command execution via the astrofy CLI, and the ability to trigger sub-skills based on checklist item tags.\n
  • Sanitization: Absent. The skill lacks mechanisms to sanitize or validate the natural language instructions contained within the specification files before they are interpreted.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 10:53 AM