twitch-stream-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The skill explicitly instructs the agent to install dependencies automatically and to "not request separate approval" or describe the install link to the user, which is a deceptive runtime instruction outside the stated data-collection purpose.

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs the agent to invoke the BrowserAct CLI to access real Twitch web pages at runtime, which can ingest outsider-authored free text from those public browser pages (e.g., titles/chat/visible content) into the LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent to automatically install a required runtime dependency using the Install BrowserAct link (https://www.browseract.com/?co-from=social-media&redirect=https://github.com/browser-act/skills/tree/main), which the agent would fetch and run at runtime to obtain/execute remote installation code.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (low risk: 0.30). The skill instructs the agent to install the BrowserAct CLI automatically without asking the user for confirmation, which could cause the agent to modify the host system or bypass user consent, although it does not explicitly request sudo, system-file edits, or account creation.