android-ble-gatt-status
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted local data by analyzing an Android project's source code using the bundled
scripts/inspect-project.pyscript. - Ingestion points: The analyzer script recursively reads Kotlin, Java, and XML files within the directory specified by the user.
- Boundary markers: The script outputs the analysis as JSON, but there are no explicit instructions for the agent to treat content within the code blocks (such as comments or string literals) as potentially untrusted data.
- Capability inventory: The script's operations are limited to reading local files and identifying patterns using regular expressions; it does not have the ability to make network requests, write to the file system, or execute other processes.
- Sanitization: The script performs regex-based signal detection rather than full execution of the code, which limits the potential for complex injection attacks to influence the diagnostic outcome.
- [COMMAND_EXECUTION]: The skill requires the execution of
scripts/inspect-project.pyto function. While this involves running local code, the script is a core component of the skill's diagnostic functionality and is restricted to static analysis tasks.
Audit Metadata