doc-stest-validator

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Python and Bash scripts to validate documentation artifacts. These scripts are located within the ai_dev_ssd_flow project structure.\n- [PROMPT_INJECTION]: The skill analyzes local files, which serves as an ingestion point for potentially untrusted content. This creates a surface for indirect prompt injection, although no specific exploitation logic was found.\n
  • Ingestion points: Artifact files located in docs/10_TSPEC/STEST/ as defined in SKILL.md.\n
  • Boundary markers: No specific boundary markers are utilized in the skill instructions.\n
  • Capability inventory: Commands execute python and bash scripts on local files as part of the validation process.\n
  • Sanitization: There is no evidence of content sanitization within the skill configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 12:13 PM