The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is instructed to load and follow instructions from external documentation files (docs/features/<key>/intent-<key>.md) which can override the agent's default behavior.
Ingestion points: Reads docs/features/<key>/intent-<key>.md at the start of the workflow.
Boundary markers: No explicit boundary markers or 'ignore' instructions are provided when interpolating these files into the context.
Capability inventory: The skill has access to high-privilege tools including Bash, Edit, Write, and Skill (recursive skill execution).
Sanitization: There is no evidence of sanitization or validation of the content within the intent files.
[COMMAND_EXECUTION]: The skill executes a local maintenance script via Node.js at startup.
Evidence: node "${CLAUDE_PLUGIN_ROOT}/skills/agentctl-setup/scripts/agentctl-setup.js" status
Context: The script resides within the plugin root directory and appears to be a status check for a helper tool, which is common in development-focused skills.

feature-dev