starry-slides

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to "Load the remote contract, CLI usage, and discovery references" from public raw.githubusercontent.com URLs (the Starry Slides contract, CLI usage, and discovery interview), which are untrusted third‑party documents the agent must read and that can materially influence generation and actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs the agent to load and use remote authoritative documents at runtime from raw.githubusercontent.com (https://raw.githubusercontent.com/StarryKit/starry-slides/main/docs/skills-references/STARRY-SLIDES-CONTRACT.md, https://raw.githubusercontent.com/StarryKit/starry-slides/main/docs/skills-references/STARRY-SLIDES-CLI-USAGE.md, https://raw.githubusercontent.com/StarryKit/starry-slides/main/docs/skills-references/REQUIREMENTS-DISCOVERY-INTERVIEW.md), and those fetched documents are treated as the authoritative instructions that directly control prompt generation and agent behavior.