The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the vendor's langwatch CLI tool to perform prompt management operations such as prompt init, prompt create, and prompt sync.
[EXTERNAL_DOWNLOADS]: The agent is instructed to fetch documentation directly from the vendor's official website (langwatch.ai) and via the CLI tool to ensure up-to-date integration guidance.
[DATA_EXFILTRATION]: Includes a diagnostic reporting feature via npx langwatch report. This process is designed with safety in mind, requiring explicit user approval (--user-approved) and performing local scrubbing of sensitive data (secrets, PII, API keys) before transmission to the vendor.
[CREDENTIALS_UNSAFE]: Instructs the agent to retrieve the LANGWATCH_API_KEY from the project's .env file or environment variables. The instructions include a critical safety warning to avoid reading or exposing unrelated credentials present in the same file.

prompts