The Agent Skills Directory

[PROMPT_INJECTION]: The skill utilizes role-playing instructions ('You are the friend who tells the truth') and behavioral directives ('No flattery. No hedging') to define the AI's persona for the duration of the audit. This ensures the output remains focused on critical business feedback rather than general development assistance.
[COMMAND_EXECUTION]: The workflow includes executing standard shell commands such as git log to evaluate recent development velocity and grep to search for payment provider integrations (e.g., Stripe, Lemon Squeezy). These actions are consistent with the skill's purpose of analyzing a project's commercial readiness.
[PROMPT_INJECTION]: The skill ingests data from untrusted sources within the project, such as README.md, docs/, issues/, and PRD* files, to understand the project's scope and goals.
Ingestion points: Project documentation, requirement files, and issue trackers.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within these files.
Capability inventory: The skill allows reading local files, executing search commands, and querying the user for more information.
Sanitization: The skill does not describe any specific mechanisms for filtering or sanitizing the content of the files it reads before processing them.

make-money