techsmith-reference-architecture

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 local project files (scripts, manifests, logs) and fetches remote documentation from TechSmith servers. While this creates a potential surface for indirect prompt injection if those files contain malicious instructions, the skill includes explicit instructions to redact sensitive data and verify approval boundaries before performing write operations.
  • Ingestion points: Local script discovery (Read, Glob, Grep) and remote documentation fetching (WebFetch targeted at TechSmith documentation domains).
  • Boundary markers: Explicit instructions to confirm the target repository file and approval boundaries before using Write or Edit tools.
  • Capability inventory: The skill is restricted to standard file access (Read, Write, Edit, Glob, Grep) and web fetching (WebFetch). It does not have access to shell execution or privileged system tools.
  • Sanitization: Instructions mandate the redaction of all keys, account identifiers, activation artifacts, and sensitive endpoint details.
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches configuration and documentation from official TechSmith domains (support.techsmith.com, assets.techsmith.com) and their official GitHub organization (github.com/TechSmith). These are well-known, legitimate sources for product documentation and samples related to the skill's primary purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 04:29 PM
Security Audit — agent-trust-hub — techsmith-reference-architecture