socratic-method
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests historical learning data from a local file to adjust agent behavior and questioning rigor.
- Ingestion points:
~/.junior-growth/mastery.mdis read at the start of tasks to check mastery status for relevant concepts. - Boundary markers: No specific delimiters or "ignore instructions" markers are used when reading this tracking file.
- Capability inventory: The skill allows the agent to update the mastery file at the end of different task phases and gates technical responses based on the file's content.
- Sanitization: The skill does not explicitly sanitize the markdown content in the mastery file, which is created and maintained by the agent itself.
- [SAFE]: The skill follows established development practices, including manual user-driven setup for global application and maintaining task-specific state locally without external network transmission. All documented setup steps involve standard symbolic linking and user-controlled configuration files.
Audit Metadata