gplay-preflight
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves reading output from the
gplay preflighttool, which decodesAndroidManifest.xmland other metadata from untrusted.apkor.aabfiles. The agent is specifically instructed to report fields likehint,message, andreffrom this output. If a malicious artifact contains specially crafted strings in its manifest, these could be reflected in the tool's output and potentially influence the agent's behavior. - Ingestion points: The
JSON outputfromgplay preflight --output json(mentioned inSKILL.md) is the primary source of untrusted data. - Boundary markers: The skill lacks explicit boundary markers or instructions for the agent to treat the tool output as potentially adversarial data.
- Capability inventory: The agent is authorized to run the
gplay preflightcommand and is integrated into a workflow that includes "upload or release" actions, providing a path for potential multi-step exploitation. - Sanitization: There are no instructions for sanitizing or escaping the content retrieved from the scanner's report before it is processed by the agent.
Audit Metadata