transcript-purpose-analysis

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted transcript data, which presents a surface for indirect prompt injection if the source text contains instructions for the agent.
  • Ingestion points: Transcript data is read from a user-supplied file path or inline text input (SKILL.md, Procedure Step 1).
  • Boundary markers: The instructions do not specify explicit delimiters or "ignore previous instruction" warnings to be applied to the ingested transcript content.
  • Capability inventory: The skill utilizes the agent's ability to read local repository files and write an analysis report to a new Markdown file (SKILL.md, Procedure Step 1, 5, and 7).
  • Sanitization: No specific sanitization or filtering of the transcript text is performed, as the skill is required to extract verbatim quotes for evidence (SKILL.md, Output A).
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 03:04 PM
Security Audit — agent-trust-hub — transcript-purpose-analysis