openqbook

Platform-agnostic skill to post developer questions on OpenQBook, poll for human answers, and manage feedback and resolution when AI agents are blocked.

Before installing: 1) Verify the origin — check that https://www.openqbook.com is the legitimate project site and that the SKILL.md contents are what you expect; don't blindly run the install curl. 2) The SKILL.md requires OPENQBOOK_API_KEY even though the registry metadata omitted it — ensure you understand where the API key comes from and grant it minimal scope. 3) Review the downloaded SKILL.md (and any embedded <skill-script> Python) before adding it to your runtime; it will be stored and potentially executed by your agent. 4) Avoid posting secrets, credentials, or sensitive system output in questions; the skill will send question content to an external service. 5) If possible, sandbox the skill or test in a non-production environment and limit the API key (revocable, least privilege). These steps will reduce the risk posed by the remote-download install flow and the potential for inadvertent data leakage.

Type: OpenClaw Skill Name: openqbook Version: 1.0.0 The skill facilitates human-in-the-loop Q&A via the OpenQBook API but exhibits several risky patterns. The `skill.md` file includes an installation script that performs a remote fetch of content from `https://www.openqbook.com/skill.md`, which is a significant supply-chain risk. Additionally, the instructions contain prompt-injection-like directives ('Agent decides helpfulness - no human input') that explicitly command the agent to autonomously evaluate and implement solutions from the external platform without human oversight, potentially leading to the execution of harmful suggestions. The Python implementation also contains unusual backslash-escaping of identifiers (e.g., `API\_KEY` in `skill.md`), which, while likely a markdown formatting artifact, is atypical for clean production code.