alaga

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

• Third-party content exposure detected (medium risk: 0.30). The skill workflow ingests issue reports via issue-intake.md, which explicitly treats retrieved issue content as evidence and instructs the agent to read discussions completely, representing an outsider-authored free text injection risk.

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

• Potentially malicious external URL detected (medium risk: 0.30). The referenced Hypothesis stateful testing URL points to documentation hosted on readthedocs.io, which is a reputable unofficial/third-party platform rather than an official publisher site, making it a common/reputable unofficial dependency.