The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several shell commands for codebase discovery, including find, grep, cat, and ls. It also executes framework-specific commands such as php artisan route:list to audit application accessibility and navigation paths.
[INDIRECT_PROMPT_INJECTION]: The skill ingests a large volume of application source code and configuration files, creating a surface where malicious instructions embedded in the codebase could influence the agent's behavior.
Ingestion points: Processes files across the app/, config/, database/, and routes/ directories during the "Deep Discovery" phase.
Boundary markers: The skill does not implement delimiters or explicit instructions to distinguish analyzed code from agent directives.
Capability inventory: The skill can write to the local filesystem (creating docs/detailed-test-list.md) and execute shell commands for analysis.
Sanitization: No filtering or sanitization is performed on the file content before it is read into the agent's context.

test-plan