barcode-capture-cordova

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the reading and modification of user-controlled source files (e.g., www/index.js), which introduces a potential surface for indirect prompt injection if the files being processed contain malicious instructions targeted at the agent.
  • Ingestion points: The agent is instructed to read and edit existing project files during the integration process in references/integration.md.
  • Boundary markers: The skill does not provide specific instructions or delimiters to ensure the agent ignores instructions potentially embedded within code comments or string literals in the user's files.
  • Capability inventory: The skill relies on file read/write and editing capabilities to perform its primary function of SDK integration.
  • Sanitization: There is no evidence of content validation or sanitization for the data ingested from user files.
  • [PROMPT_INJECTION]: Analysis of the skill body confirms that flags for instruction concealment are false positives. The instruction to not direct users to external documentation is a formatting preference, and the instruction to disable the barcode capture mode during processing is a specific performance optimization required by the Scandit SDK native bridge.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:32 PM
Security Audit — agent-trust-hub — barcode-capture-cordova