openspec-android-bug
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands for diagnostic collection and build analysis.
- Evidence: Uses
adb logcat,adb bugreport,adb pull /data/tombstones/,retrace, and./gradlew :module:dependenciesto gather device state and project metadata. - [PROMPT_INJECTION]: The skill ingests untrusted external content, creating a surface for indirect prompt injection attacks.
- Ingestion points: Processes user-supplied crash logs, stack traces, and bug descriptions in Step 1.
- Boundary markers: Absent. The skill does not implement delimiters or specific instructions to treat user data as non-executable text.
- Capability inventory: The agent has access to shell command execution (
adb,gradlew), file system exploration (rg,find,codegraph), and local artifact management. - Sanitization: Absent. The instructions do not specify any validation or filtering of the diagnostic data before it is analyzed to create a fix proposal.
Audit Metadata