accesslint-scan

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

• Third-party content exposure detected (high risk: 0.85). The required workflow runs npx -y @accesslint/cli@latest "<url>" ... --format json, which fetches and audits a live page from a user-supplied URL at runtime; the page’s DOM/text is ingested by the accesslint tooling and then returned as JSON that the agent reads into LLM context, making it outsider-authored free text from arbitrary web content.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill invokes npx to fetch-and-run remote npm packages (@accesslint/chrome@latest and @accesslint/cli@latest) from the npm registry (e.g. https://registry.npmjs.org/@accesslint/chrome and https://registry.npmjs.org/@accesslint/cli) at runtime, which executes remote code and is required for the skill to operate.