The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown files containing business advice, personality profiles, and routing instructions. No executable scripts, configuration files for package managers, or binary assets are present.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user questions. It correctly implements boundary markers by instructing the agent to 'answer in that expert's voice using only the substance in the files you loaded,' which restricts the model from following external instructions embedded in user queries. Given that the skill has no capability to write to the filesystem, access the network, or execute shell commands, the risk associated with this attack surface is negligible.

dojo-keith-rabois