lark-unified

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md and reference docs explicitly instruct the agent to fetch and process user-generated/untrusted content from Lark/Feishu (e.g., "lark-cli im +messages-search", "lark-cli doc +docs-get", "lark-cli mail +messages-search", "lark-cli wiki spaces get_node" and file downloads in references/lark-im.md, references/lark-doc.md, references/lark-mail.md), so third-party content is ingested and can materially influence actions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's setup step can run "npm install -g @larksuite/cli" at runtime, which fetches and installs/executes remote code from the npm registry (e.g. https://registry.npmjs.org/@larksuite/cli) and the skill depends on lark-cli to operate, so this is a runtime external dependency that executes remote code.