techsmith-core-workflow-a

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external inputs such as capture subjects and destination paths, which serves as a potential vector for instructions embedded in data.
  • Ingestion points: The skill accepts a [script-path] and [approved-output-directory] via arguments and processes user-defined capture subjects.
  • Boundary markers: Instructions explicitly mandate path validation, canonicalization of the LastFileWritten property, and checks to ensure the output path does not escape the approved boundary.
  • Capability inventory: The skill uses Write and Edit tools alongside the Snagit.ImageCapture.1 COM interface to generate and manage image files.
  • Sanitization: The contract requires redacting all activation keys, account identifiers, and sensitive endpoint details before returning output.
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches content from official TechSmith infrastructure and repositories.
  • Evidence: References documentation and PDF guides from assets.techsmith.com and support.techsmith.com.
  • Evidence: Directs users to official PowerShell samples hosted in the TechSmith GitHub organization.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 04:28 PM
Security Audit — agent-trust-hub — techsmith-core-workflow-a