The Agent Skills Directory

[SAFE]: The skill provides a TypeScript wrapper for the official Anthropic SDK to facilitate structured logging. This is a standard observability pattern and does not involve any malicious data exfiltration or unauthorized access.- [SAFE]: All external links point to official Anthropic domains (console.anthropic.com and docs.anthropic.com), which are recognized as trusted services.- [INDIRECT_PROMPT_INJECTION]: The trackedCreate function facilitates interactions with an LLM by wrapping request parameters. As with any LLM integration, developers should ensure that user-controlled data within these parameters is appropriately handled, though this skill does not introduce any specific vulnerabilities or bypasses.

clade-observability