notebook-for-experiment

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). In scripts/nb.py, commands like set-cell/insert-cell read attacker-controlled free text from stdin or a user-supplied file path and write it into the notebook cell source, which is then later consumed by execute/export-html (i.e., the workflow ingests arbitrary outsider-authored text at runtime without selecting a specific trusted item).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The SKILL includes setup commands that fetch-and-execute remote install scripts (curl ... | sh and irm ... | iex) from https://astral.sh/uv/install.sh and https://astral.sh/uv/install.ps1, which clearly execute remote code on the host.