skills/open-gsd/gsd-core/gsd-settings/Gen Agent Trust Hub

gsd-settings

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by reading and parsing existing configuration files, which could contain injected instructions that influence agent behavior. \n
  • Ingestion points: Reads current settings from the local file ~/.claude/gsd-core/workflows/settings.md (SKILL.md). \n
  • Boundary markers: No explicit boundary markers or 'ignore' instructions are provided to the agent to isolate the parsed configuration data from the system prompt. \n
  • Capability inventory: The skill has access to Bash, Read, and Write tools, representing a high capability tier for potential exploits (SKILL.md). \n
  • Sanitization: The instructions do not specify any validation or sanitization routines for the configuration data before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 12:43 AM
Security Audit — agent-trust-hub — gsd-settings