The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses dgrunner, docker, and npm CLI tools to perform runner lifecycle operations (start, stop, restart, remove) and diagnostics. These are standard administrative functions for managing a self-hosted runner.
[INDIRECT_PROMPT_INJECTION]: The skill provides an interface to read logs from the runner process (dgrunner logs --follow) and Docker containers. This introduces a surface where data from the runner's execution environment is ingested into the agent's context.
Ingestion points: Log output streams from dgrunner logs and docker logs as described in SKILL.md.
Boundary markers: No specific delimiters or instructions are provided to the agent to treat log content as untrusted data.
Capability inventory: The agent can execute runner management commands, modify the installation directory via --dir, and update the tool via npm.
Sanitization: No sanitization or filtering of the streamed log content is implemented.

dgrunner-operations