barcode-capture-ios
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill contains instructions that prioritize provided reference files over the model's internal training data to ensure API accuracy. These are constructive behavioral constraints for developer documentation rather than malicious injections.
- [CREDENTIALS_UNSAFE]: The skill provides clear guidance on handling Scandit license keys. It explicitly instructs the agent to mask keys in the chat interface and prohibits searching the project's files or environment for keys, which is a defensive security measure to prevent accidental exposure.
- [EXTERNAL_DOWNLOADS]: All external URLs referenced target official Scandit domains (docs.scandit.com, ssl.scandit.com) or their official GitHub repositories. These are trusted vendor resources used for documentation and SDK distribution via Swift Package Manager.
- [COMMAND_EXECUTION]: The skill instructs the agent to write integration code into user files upon request. This is standard behavior for a coding assistant skill and does not involve the execution of arbitrary or dangerous system commands.
- [INDIRECT_PROMPT_INJECTION]: The skill has a minimal attack surface for indirect injection as it primarily generates code from static markdown templates. It includes instructions to validate API usage against provided references, which helps mitigate the impact of potentially poisoned external data.
Audit Metadata