The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is built around the execution of the asc CLI tool to interact with App Store Connect iris (private) API endpoints.
[CREDENTIALS_UNSAFE]: The skill documents the use of sensitive browser session cookies (myacinfo, itctx, dqsid, wosid) for authentication. It provides instructions for manual entry via the ASC_IRIS_COOKIES environment variable or automated extraction from local browser databases.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to use the affordances field in JSON responses to find and execute next-step commands.
Ingestion points: Data returned from shell commands such as asc iris status and asc iris apps list (as described in SKILL.md).
Boundary markers: No specific delimiters or warnings are provided to differentiate between data and potentially malicious instructions within these affordances.
Capability inventory: Full access to the asc CLI tool for creating apps, listing data, and registering bundle identifiers.
Sanitization: The skill does not describe any validation or sanitization of the commands suggested in the affordances field before the agent executes them.

asc-iris