tinyfish-search

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path “Fetch Content” calls tinyfish fetch content get "<url>" --format markdown, which ingests markdown scraped from a user-supplied URL (public web content authored by outsiders) into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly runs runtime fetches of arbitrary external pages via the TinyFish CLI (e.g., "tinyfish fetch content get "" --format markdown"), which will inject fetched remote content into the agent's context and thus can directly control prompts/outputs.