The Agent Skills Directory

[DATA_EXFILTRATION]: The skill communicates with well-known AI provider APIs (Anthropic, OpenAI, Perplexity, and OpenRouter) to measure brand visibility through a series of strategic queries. These operations are essential to the skill's function and utilize environment variables for authentication, adhering to secure credential management practices.
[EXTERNAL_DOWNLOADS]: The Python scripts citability_scorer.py and schema_validator.py fetch HTML content and JSON-LD data from external websites specified by the user. These network requests are necessary for performing site audits and use standard libraries without suspicious download-and-execute patterns.
[COMMAND_EXECUTION]: The repository includes a push.sh helper script designed for repository maintenance. This script executes standard git commands and incorporates a proactive security check to prevent the accidental commitment of API keys to the version control system.
[SAFE]: No evidence of prompt injection, persistence mechanisms, obfuscation, or unauthorized privilege escalation was found. The skill's behavior is transparent, well-documented, and strictly aligned with its stated analytical purpose.

aeo-audit