The Agent Skills Directory

[CREDENTIALS_UNSAFE] (HIGH): The skill explicitly lists sensitive environment variables used for App Store Connect authentication, such as ASC_PRIVATE_KEY, ASC_PRIVATE_KEY_B64, ASC_KEY_ID, and ASC_ISSUER_ID. This documentation provides a roadmap for an attacker to target specific secrets for exfiltration via prompt injection or environment dumping.
[COMMAND_EXECUTION] (MEDIUM): The skill's primary purpose is to instruct the agent to execute system commands (asc). This introduces risks of command injection if the agent attempts to interpolate untrusted user input or external data into the CLI flags (e.g., app names or build IDs).
[INDIRECT_PROMPT_INJECTION] (HIGH): The skill is designed to process external data returned from the App Store Connect API.
Ingestion points: CLI output from asc (fetched from Apple's servers).
Boundary markers: Absent. There are no instructions to treat CLI output as untrusted data or use delimiters.
Capability inventory: The skill possesses the capability to execute system commands and access sensitive authentication tokens.
Sanitization: None. The agent is encouraged to parse various output formats (JSON, Table, Markdown) directly.
[EXTERNAL_DOWNLOADS] (LOW): While the skill does not provide a download command, it assumes the presence of a third-party CLI tool (asc) that is not part of the standard environment, representing an unverifiable dependency.

asc-cli-usage