The Agent Skills Directory

[SAFE]: The skill provides standard architectural patterns for observability, including structured JSON logging, health check endpoints (liveness and readiness), and request middleware for correlation IDs.
[DATA_EXPOSURE]: The skill explicitly includes a coordination rule to avoid logging sensitive data such as passwords, tokens, or PII.
[COMMAND_EXECUTION]: While the skill frontmatter allows the use of the Bash tool, the provided examples and instructions focus on standard application instrumentation and Docker health check configurations rather than arbitrary command execution.
[INDIRECT_PROMPT_INJECTION]: The skill processes internal build contracts and service maps to understand the infrastructure. It includes structured coordination rules to prevent the agent from modifying business logic, effectively bounding its operations to instrumentation directories (src/telemetry/, src/logging/).

observability-agent