The Agent Skills Directory

[SAFE]: The skill's content is entirely instructional and documentation-based, providing architectural guidance for observability. It references official and well-known documentation from authoritative sources like the W3C, Google SRE, and OpenTelemetry. No signs of obfuscation, exfiltration, or unauthorized command execution were found.- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the analysis of user-provided implementation plans and code snippets. While this constitutes an indirect prompt injection surface, the skill lacks tool-based capabilities (such as shell access or network operations) to exploit this surface. Furthermore, the instructions explicitly guide the agent to review for log injection risks and enforce the redaction of sensitive credentials and PII, enhancing the overall security posture of the workflows it assists with.

production-observability