text-trainer
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Vulnerability to indirect prompt injection. The skill is designed to ingest and process untrusted writing samples from users (via chat, files, or directories) as the primary input for its analysis engine.
- Ingestion points: SKILL.md (Step 2) defines methods for collecting external text data from users and local filesystems.
- Boundary markers: The instructions do not employ delimiters or specific commands for the agent to ignore any instructions found within the processed samples.
- Capability inventory: The agent has filesystem read and write capabilities, including reading from .agents/gtm-context.md and writing to .agents/voice-profiles/ (SKILL.md Step 0, Step 5).
- Sanitization: There is no evidence of filtering or sanitization of the input text before it is analyzed by the 8-pass framework.
Audit Metadata