ab-tests
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONPRIVILEGE_ESCALATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill setup routine attempts to download the adapty CLI package from the official NPM registry if it is not found on the system.
- [REMOTE_CODE_EXECUTION]: Uses npx --yes adapty@latest to execute remote code from the NPM registry, allowing the agent to run the most recent version of the vendor tool without manual installation.
- [COMMAND_EXECUTION]: Relies on executing shell commands through a resolved CLI path to interact with the Adapty API for operations including ab-tests, paywalls, placements, and flows management.
- [PRIVILEGE_ESCALATION]: The setup script includes an attempt to install the package globally using npm i -g, which typically triggers a request for elevated system permissions.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external API outputs (such as placement configurations and segment lists) to generate user-facing summary cards.
- Ingestion points: Data returned from placements get, ab-tests get, paywalls list, and flows list (SKILL.md).
- Boundary markers: Not explicitly defined in the data interpolation logic, though instructional guidelines advise the agent on how to present the information.
- Capability inventory: Significant capabilities including ab-tests start, stop, replace, archive and placements/paywalls create (SKILL.md).
- Sanitization: The skill emphasizes privacy for outbound feedback notes but does not specify sanitization or validation of the structured JSON data received from the API before processing.
Audit Metadata