AGENT LAB: SKILLS

meeting-minutes

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION] (LOW): The skill is vulnerable to indirect prompt injection because it is designed to process untrusted data from meeting transcripts and notes. * Ingestion points: Operational workflow and example prompts in SKILL.md refer to transcripts and raw notes. * Boundary markers: Absent; there are no instructions to use delimiters or ignore instructions within transcripts. * Capability inventory: None; the skill lacks code execution, file system, or network tools. * Sanitization: Absent; no filtering of input text is defined.
  • [NO_CODE] (SAFE): The skill contains only markdown instructions and no executable scripts, binaries, or remote dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:31 PM