sosumi

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md explicitly instructs the agent to fetch and read public third‑party web content (e.g., via the fetchExternalDocumentation tool and the https://sosumi.ai/external/{full-https-url} path and examples like https://apple.github.io/... and Swift Package Index pages) and to use that Markdown content when answering implementation questions, so untrusted/user-hosted pages can materially influence agent decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). I flag https://sosumi.ai/mcp because the skill requires connecting to that external MCP endpoint at runtime to fetch documentation/transcripts which are injected into the model context and therefore directly influence agent prompts and behavior.