The Agent Skills Directory

[SAFE]: The skill provides purely instructional content and documentation for interacting with the Goldsky Edge API. No executable code or suspicious automation scripts are included in the skill body.
[EXTERNAL_DOWNLOADS]: The skill references official documentation and configuration endpoints such as edge.goldsky.com and docs.erpc.cloud/llms.txt. These resources are intended to provide the AI agent with additional context and registry information for legitimate service operation. These are documented neutrally as standard service interactions.
[CREDENTIALS_UNSAFE]: The skill correctly uses placeholders like YOUR_SECRET and gs_edge_... to demonstrate authentication patterns. It explicitly advises users to use HTTP headers (X-ERPC-Secret-Token) rather than query parameters to prevent secret leakage in logs, demonstrating a follow-through of security best practices.
[INDIRECT_PROMPT_INJECTION]: The skill documents an interface for processing blockchain RPC data. While this represents a data ingestion surface, the instructions are focused on standard API usage, and there are no patterns suggesting exploitation of the agent's processing logic. The risk is negligible and consistent with the skill's primary purpose.

edge