barcode-capture-web

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and downloads official Scandit SDK components from the well-known jsDelivr CDN (https://cdn.jsdelivr.net/npm/@scandit/web-datacapture-barcode@8/sdc-lib/). This is a standard practice for web-based library integration.
  • [PROMPT_INJECTION]: The static analysis flagged instructions directing the agent not to refer users to external documentation. In context, this is a benign instruction aimed at ensuring the agent provides direct answers using the skill's reference files, rather than a malicious attempt to conceal behavior or bypass safety filters.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of data from scanned barcodes, which constitutes untrusted external input.
  • Ingestion points: Barcode data is ingested via the didScan event handler defined in references/integration.md.
  • Boundary markers: The skill provides strict API usage guidelines and explicit code templates to manage data flow.
  • Capability inventory: The functionality is confined to standard browser-based barcode capture, UI updates, and console logging. No dangerous capabilities such as file-system writes, shell command execution, or network exfiltration to unknown domains are present.
  • Sanitization: Scanned data is handled as standard strings for UI and logging purposes without being passed to sensitive sinks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:32 PM
Security Audit — agent-trust-hub — barcode-capture-web