The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted content from project-specific documents and deliverables, which creates an indirect prompt injection surface.
Ingestion points: The skill reads stakeholder prose from scope/key-question.md and deliverable artifacts located in action-fields/ directories.
Boundary markers: There are no explicit markers or instructions to isolate the ingested project data from the instructions provided to the agent or the persona challenger agent.
Capability inventory: The skill has the ability to execute internal bash scripts, modify files via Write and Edit, and dispatch other skills/agents.
Sanitization: No sanitization or validation of the ingested text is performed before it is used to define personas or generate critiques.
[COMMAND_EXECUTION]: The skill uses the Bash tool to run administrative scripts and manage files within its environment.
It executes a project discovery script (discover-projects.sh) and uses standard shell commands like cp to initialize persona templates from its internal reference directory. These operations are limited to the plugin's own filesystem structure.

consult-personas