The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary function is to execute shell commands in the background using the 'bgproc' utility (e.g., bgproc start -n name -- command). This is the intended purpose of the process management tool.
[EXTERNAL_DOWNLOADS]: The skill metadata specifies the installation of the 'bgproc' package from the public NPM registry using bun i -g bgproc.
[DYNAMIC_EXECUTION]: The skill facilitates the dynamic execution of subprocesses by allowing the agent or user to specify arbitrary commands to be run and managed in the background.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data through the bgproc logs command, which reads the output of managed processes.
Ingestion points: Process logs are read from the data directory via bgproc logs (documented in SKILL.md).
Boundary markers: The skill lacks explicit boundary markers for prompt injection but includes a proactive security warning to redact secrets from the log streams.
Capability inventory: The skill can start subprocesses (bgproc start), restart them, and view logs across the system (on macOS/Linux).
Sanitization: No automatic sanitization of log content is described, relying on the user/agent to redact sensitive information manually.

jb-bgproc