The Agent Skills Directory

[SAFE]: The skill implements a mechanism to load local agent definition files from the agents/ directory on-demand. This behavior is restricted to the local file system and is consistent with the skill's stated purpose of context optimization.
[SAFE]: No evidence of prompt injection, data exfiltration, obfuscation, or remote code execution was found. The structured data blocks (labeled as toon) are used for indexing and scoring agents and do not contain executable or hidden malicious content.
[SAFE]: The use of the cat command in documentation is illustrative of how the loader accesses local files and does not constitute a risk of arbitrary command execution or privilege escalation.

lazy-agent-loader