learning-design-checklist
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads from an external data source and interpolates it into the agent's output, which is a pattern susceptible to instruction injection if the data source is untrusted.
- Ingestion points: The skill loads design principles from the local path
~/.claude/skills/learning-design-pillars/principles/learning-design-pillars.json. - Boundary markers: The instructions do not define any delimiters or clear warnings for the agent to treat the JSON content as untrusted data or to ignore instructions embedded within the text.
- Capability inventory: The skill's defined workflow is limited to reading a file and generating text output; it does not instruct the agent to use tools for network communication, file system modification, or shell command execution.
- Sanitization: No data validation, escaping, or sanitization steps are defined for the ingested JSON content before it is processed by the agent.
Audit Metadata