The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructions and documentation explicitly reference the path to a sensitive App Store Connect private key file at ~/.asc/AuthKey.p8. Instructing an agent to interact with or authenticate using private keys at specific file paths presents a high risk of credential exposure.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external sources (App Store customer reviews).
Ingestion points: Customer reviews are fetched using the asc reviews list and asc reviews get commands.
Boundary markers: Absent. The instructions do not provide delimiters or warnings to the agent to disregard potential instructions embedded within the review text.
Capability inventory: The skill has the capability to execute commands via the asc CLI, including creating and deleting developer responses.
Sanitization: None detected. There is no evidence of filtering or validation of the review content before the agent processes it or generates responses based on it.
[COMMAND_EXECUTION]: The workflow involves executing shell commands using the asc CLI and piping data to jq. This includes reading local configuration files like .asc/project.json and dynamically constructing commands, which could be misused if internal variables are manipulated.

asc-customer-reviews