The Agent Skills Directory

[DYNAMIC_EXECUTION]: The platform provides a runCode tool that executes arbitrary asynchronous TypeScript code on the Polylane backend. This enables complex data processing and tool orchestration within the vendor's environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to process data from various untrusted external sources, such as production logs, repository content, and third-party observability integrations.\n
Ingestion points: Production logs (service logs), repository files (repo read), and telemetry data from connected integrations (AWS, Sentry, Datadog).\n
Boundary markers: No specific delimiters or safety warnings are provided for the interpolation of external data into the agent's context.\n
Capability inventory: The agent has access to file system operations (via repo tools), network requests (via execute), and remote code execution (runCode).\n
Sanitization: The skill does not describe any specific sanitization or filtering protocols for external data before analysis.\n- [COMMAND_EXECUTION]: The skill defines patterns for using the polylane CLI and specialized MCP tools to manage cloud infrastructure, deployments, and incident triage.\n- [EXTERNAL_DOWNLOADS]: The skill fetches documentation, configuration templates, and setup instructions from official vendor domains, including docs.polylane.com and api.polylane.com.

polylane