humanize-text
Warn
Audited by Gen Agent Trust Hub on Jun 25, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to install a plugin from an unverified community repository (github.com/gregorymm/humanize-text) using commands like /plugin install humanize-text@gregorymm-humanize.
- [PROMPT_INJECTION]: The skill is designed to ingest untrusted data from multiple sources (text, Figma URLs, and screenshots), creating an attack surface for indirect prompt injection where instructions hidden in the data could be executed by the agent.
- Ingestion points: Raw text, Figma URLs, and UI screenshots (specified in SKILL.md).
- Boundary markers: No delimiters or instructions to ignore embedded commands are provided in the instructions.
- Capability inventory: No executable scripts or local code are provided in this catalogue stub, which limits the immediate risk of the stub itself.
- Sanitization: No evidence of input validation or sanitization is present in the documentation.
Audit Metadata