monte-carlo-instrument-agent

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly runs scripts/fetch_sdk_docs.py to fetch the Monte Carlo SDK's supported-instrumentor list live from PyPI (https://pypi.org/project/montecarlo-opentelemetry/) and the workflow uses that live PyPI README data to choose/install instrumentors and version pins, so untrusted third-party content on PyPI is read and directly influences installer/diff decisions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill runs scripts/fetch_sdk_docs.py at runtime to query PyPI (https://pypi.org/project/montecarlo-opentelemetry/), and the live PyPI response is required and directly controls which instrumentor packages and version pins the skill proposes (thus influencing generated setup/install instructions), so this external fetch is a runtime dependency that controls agent instructions.