The Agent Skills Directory

[Indirect Prompt Injection] (HIGH): The skill's purpose is to process and validate external data (schemas, ETL patterns), while simultaneously requesting high-privilege tools including 'Bash', 'Write', and 'Edit'. This combination creates a high-risk capability tier where malicious instructions embedded in a schema could be executed by the agent.
Ingestion points: Data pipeline schemas and transformation request inputs.
Boundary markers: Absent; no instructions provided to delimit or ignore instructions within data.
Capability inventory: 'Bash', 'Write', 'Edit', 'Read', and 'Grep'.
Sanitization: None; the skill lacks logic to sanitize or validate the safety of the content it processes.
[Command Execution] (MEDIUM): The skill requests 'Bash' tool access. For a schema validation task, providing shell access violates the principle of least privilege and increases the impact of any successful injection.
[No Code] (INFO): The skill consists only of metadata and descriptions in a markdown file. No actual implementation logic or scripts are provided, preventing a complete review of its operational behavior.

schema-validator