asc-submission-health
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes administrative commands via the asc CLI to manage App Store Connect resources, including checking build status (asc builds info), validating app readiness (asc validate), and cancelling active submissions (asc submit cancel).
- [PRIVILEGE_ESCALATION]: The skill employs authenticated web-session commands (asc web) to perform management tasks that exceed the capabilities of the public App Store Connect API, specifically for managing App Privacy publish states and app availability records.
- [CREDENTIALS_UNSAFE]: The skill manages authentication through environment variables (ASC_*) and mentions a keychain bypass flag (ASC_BYPASS_KEYCHAIN=1) for isolated environments, while including explicit instructions to avoid printing secrets in logs or handoff text.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external diagnostic data and local configuration files to determine repair strategies.
- Ingestion points: Diagnostic data from asc validate and asc status, and local files such as privacy.json or metadata directories.
- Boundary markers: The agent is instructed to report evidence and proposed repairs for user review before execution, and to hand off healthy versions to a separate skill for final submission.
- Capability inventory: Actions include metadata modification, submission cancellation, and privacy state updates.
- Sanitization: Security relies on human oversight, dry-run validation, and manual review of generated plan files before changes are applied.
Audit Metadata