The Agent Skills Directory

[SAFE]: The skill instructions and associated reference documents focus entirely on distributed systems design, providing best practices for service decomposition, communication, and resilience.
[REMOTE_CODE_EXECUTION]: No remote code execution or suspicious script downloads were detected. Code examples use standard, well-known libraries (such as pybreaker and OpenTelemetry) for illustrative purposes within a system design context.
[CREDENTIALS_UNSAFE]: No hardcoded credentials, API keys, or sensitive secrets were found. The skill correctly advocates for externalized configuration and proper security boundaries.
[DATA_EXFILTRATION]: Network operations in code snippets are limited to architectural examples using placeholder variables (e.g., INVENTORY_URL, PAYMENT_SERVICE_URL) and do not attempt to access or exfiltrate sensitive local data.
[OBFUSCATION]: No obfuscated text, Base64-encoded executable strings, or hidden Unicode characters were identified across any of the files.
[INDIRECT_PROMPT_INJECTION]: The skill is a technical reference guide and does not ingest or process untrusted external data into prompts, resulting in a minimal attack surface for indirect injection.

microservices-architect