The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes potentially untrusted data from service logs and external documentation URLs.
Ingestion points: The list_logs tool and instructions to fetch and read content from https://render.com/docs/*.md using curl or other HTTP clients in SKILL.md and referenced files.
Boundary markers: The instructions do not specify delimiting logs or documentation content to prevent the agent from interpreting data as instructions.
Capability inventory: The agent has access to sensitive tools including query_render_postgres, list_services, and the ability to execute render ssh or other CLI commands.
Sanitization: No mention of filtering, sanitization, or safety-aware parsing for the log data or external document content is provided.
[EXTERNAL_DOWNLOADS]: The skill retrieves documentation from the official render.com domain. These are well-known service resources used to provide current platform guidance to the agent.
[COMMAND_EXECUTION]: The skill relies on the render CLI and standard networking tools like curl to perform monitoring, log retrieval, and documentation fetching, which are core to its intended functionality.

render-monitor