The Agent Skills Directory

[SAFE]: The skill provides documentation and templates for schema consistency. No malicious activities, obfuscation, or data exfiltration patterns were detected.
[COMMAND_EXECUTION]: The skill suggests the use of standard search utilities like fd and rg via bash to find and inspect schema files. These are used locally for discovery and validation purposes and are safe within this context.
[INDIRECT_PROMPT_INJECTION]: The skill processes external schema files (*.aurora.yaml), which presents a surface for indirect prompt injection if those files were to contain malicious instructions. However, because the skill instructions focus on static structural and semantic validation (e.g., checking for specific keys and naming conventions), the risk is low and is an inherent part of the skill's primary function. Mandatory Evidence Chain: (1) Ingestion points: *.aurora.yaml files. (2) Boundary markers: None. (3) Capability inventory: Read, Edit, Write, Glob, Grep, Bash, WebFetch, WebSearch, Task. (4) Sanitization: None.

aurora-schema