What problem does it solve? Expo users and AI agents lack a direct channel to report broken skills, outdated docs, or tasks where an agent repeatedly failed, and there is no transparent way to control anonymous usage telemetry. ## Core Features & Use Cases - Actionable Feedback Submission: Send categorized feedback about Expo skills, docs, MCP tools, Expo CLI, or EAS CLI via the submit-expo-feedback CLI with optional category and subject flags. - Eval Candidate Reporting: Report real-world Expo tasks an AI agent failed to complete, using a structured Task/Expected/Actual/Wrong approach/Evidence format, with mandatory user approval before submission. - Opt-in Telemetry Control: Enable, disable, or check anonymous skill usage telemetry with scripts/telemetry.cjs; telemetry is off by default and never sends from CI. - Use Case: An AI agent fails three times to configure an EAS Build and the user fixes it manually. Use this Skill to draft an eval-candidate report, show it to the user for approval, and submit it with the evals category so Expo can turn the failure into an agent evaluation. ## Quick Start Ask the agent to submit feedback that the expo-router skill was missing deep-linking context, using the expo-skill-feedback skill with the skills category.
