Meta Prompt

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION] (HIGH): High risk of Indirect Prompt Injection (Category 8). The skill is designed to ingest and process user-provided strings to generate or optimize instructions.
  • Ingestion points: User input enters the context through placeholders like [prompt], [task], and [your repeated prompt] within the SKILL.md templates.
  • Boundary markers: Absent. There are no delimiters or 'ignore embedded instruction' warnings to prevent the model from executing commands contained within the user input.
  • Capability inventory: The skill's YAML frontmatter explicitly allows the 'Edit' tool, which could be abused to modify the codebase or inject malicious skills if triggered by a poisoned input prompt.
  • Sanitization: The skill lacks any logic to sanitize, escape, or validate the content of the processed prompts.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 01:24 PM