The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses curl to interact with LangSmith API endpoints for operations including listing runs, fetching project details, and deleting datasets.
[EXTERNAL_DOWNLOADS]: Fetches data including traces, runs, and evaluation datasets from the well-known LangSmith service at api.smith.langchain.com.
[PROMPT_INJECTION]: The skill processes data from external API responses (such as LLM traces and run outputs), which presents a potential surface for indirect prompt injection.
Ingestion points: Content returned from endpoints like /api/v1/runs and /api/v1/examples.
Boundary markers: The instructions do not define specific delimiters for separating API data from instructions.
Capability inventory: The skill uses curl for network communication and file system writes to /tmp/ for payload preparation.
Sanitization: No explicit sanitization or validation of the ingested API data is performed within the skill scripts.

langsmith