The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process external data that may contain malicious instructions.
Ingestion points: The skill reads member SKILL.md files during verification and system state data via CLI tools in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore or isolate embedded instructions within the processed data.
Capability inventory: The skill can execute system commands using the agenthood utility, including status checks, verification, and state restoration (rollback) in SKILL.md.
Sanitization: No mechanisms for sanitizing or validating the content of external files before processing are defined.
[COMMAND_EXECUTION]: The skill's primary operation involves the execution of shell commands through a specialized utility.
Evidence: The instructions mandate the use of agenthood status, agenthood verify, and agenthood rollback to manage the runtime environment.

runtime-health