The Agent Skills Directory

[SAFE]: The skill is a documentation-only resource that describes design patterns for monitoring agent performance, cost, and latency. It does not contain any executable tools, scripts, or commands.
[DATA_EXFILTRATION]: No unauthorized data access or exfiltration patterns were found. The skill explicitly promotes security best practices, such as redacting PII at span-write boundaries and ensuring secrets are not transmitted to third-party monitoring services.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns, package installations, or external script downloads were detected. The Python snippet provided is a benign example of token-based cost calculation logic.
[PROMPT_INJECTION]: No prompt injection or behavior override patterns were detected. The documentation actually advises against leaking trace IDs into the model context to prevent unintended side effects like cache invalidation.

telemetry-tracing