python-notebooks-async

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
  • Metadata Poisoning (MEDIUM): In 'SKILL.md', the skill's frontmatter 'name' is 'python-notebooks-async', yet the 'Invocation Notice' section explicitly instructs the AI to inform the user it is using 'python-design-modularity'. This misleading instruction is deceptive and could be used to shadow or conflict with other skills, potentially causing logic errors in how the agent identifies its active capabilities.
  • Indirect Prompt Injection (LOW): The skill is designed to process and review code, which constitutes an untrusted data ingestion surface. While vulnerable to embedded instructions in user code, the skill itself lacks high-risk capabilities such as network exfiltration or file system modification, limiting the risk to the agent's internal reasoning and local state.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 02:54 AM