de-ai

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because its primary function is to ingest and analyze untrusted text provided by a user via the --file or --text parameters. Adversarial instructions hidden within this text could attempt to override the agent's behavior.
  • Ingestion points: Data enters the context via the --text and --file parameters defined in SKILL.md and system.md.
  • Boundary markers: Absent. The instructions do not specify any delimiters or safety directives to the agent to treat the input text as data-only.
  • Capability inventory: The agent has permissions to read files and output revised text, including the creation of new files (e.g., [original]-humanized.[ext]) as described in system.md.
  • Sanitization: The skill lacks sanitization or validation logic to filter potentially malicious instructions from the input text.
  • [NO_CODE]: This skill consists entirely of instructional markdown files (README.md, SKILL.md, system.md) and does not include any Python or Node.js code, external dependencies, or binary components.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 01:03 PM
Security Audit — agent-trust-hub — de-ai