Agent Mentor
Turn a learning goal into a course with your coding agent
Agent Mentor is a local-first learning skill for Codex, Claude Code, and opencode. Tell your coding agent what you want to learn, what you already know, and how much time you have. It turns that brief into a structured, source-backed Markdown course with guided lessons, interactive exercises, a glossary, and review activities.
Unlike a one-off chat answer, the course stays in your own workflow. You can inspect, edit, version, and keep the files locally. A bundled reader provides focused lessons, progress tracking, interactive feedback, and exports to HTML, PDF, or Anki.
Agent Mentor is built for developers, students, and curious builders who want an AI agent to teach systematically instead of returning isolated answers.
Features
• Generate a structured course from a topic, learner level, and available time
• Source-backed Markdown lessons with a glossary and inspectable references
• Interactive knowledge checks with explanatory feedback
• Live coding exercises with previews and reference answers where appropriate
• Local course files you can inspect, edit, version, and back up
• Progress tracking and review activities in the bundled reader
• Export courses to HTML, PDF, or an Anki-compatible format
• Works with Codex, Claude Code, and opencode
Use Cases
• Learn agent development through a structured, project-based course
• Turn an unfamiliar framework, API, or technical topic into a learning path
• Study documentation through guided lessons and practical exercises
• Practice coding with immediate feedback and reference solutions
• Build reusable onboarding and internal training materials
• Keep course files and learning progress for future review
Comments
Hi Product Hunt 👋 I built Agent Mentor after repeatedly using coding agents to learn unfamiliar topics. The explanations were often useful, but they ended up scattered across chat threads and were difficult to review or build on. Agent Mentor turns a topic and what you already know into a structured course stored as local Markdown. It includes linked sources, guided lessons, exercises, a glossary, and spaced-review activities. You can use the bundled local reader, edit or version the files yourself, and export the course to HTML, PDF, or Anki. Each course also records its sources and passes 20+ automated checks before delivery. It currently works with Codex, Claude Code, and opencode. This is my first public launch, and I’d especially value feedback on course quality, the reader experience, and what workflow or coding agent you’d like supported next. Thanks for taking a look!
Keeping the course as inspectable, versionable Markdown is a meaningful improvement over lessons trapped in chat history. For fast-moving frameworks, it would be valuable if each course recorded source retrieval dates and dependency versions, then offered a diff-based refresh that preserves completed exercises and learner notes. A short outcome rubric before generation could also help the agent choose exercises that demonstrate mastery rather than only covering topics.
Agent Mentor Your product has strong potential, but I found a few key improvements that could make it even better. I'd love to share my feedback and suggestions—please contact me at [email protected].
Premium Products
Sponsors
Buy
Comments
Hi Product Hunt 👋 I built Agent Mentor after repeatedly using coding agents to learn unfamiliar topics. The explanations were often useful, but they ended up scattered across chat threads and were difficult to review or build on. Agent Mentor turns a topic and what you already know into a structured course stored as local Markdown. It includes linked sources, guided lessons, exercises, a glossary, and spaced-review activities. You can use the bundled local reader, edit or version the files yourself, and export the course to HTML, PDF, or Anki. Each course also records its sources and passes 20+ automated checks before delivery. It currently works with Codex, Claude Code, and opencode. This is my first public launch, and I’d especially value feedback on course quality, the reader experience, and what workflow or coding agent you’d like supported next. Thanks for taking a look!
Keeping the course as inspectable, versionable Markdown is a meaningful improvement over lessons trapped in chat history. For fast-moving frameworks, it would be valuable if each course recorded source retrieval dates and dependency versions, then offered a diff-based refresh that preserves completed exercises and learner notes. A short outcome rubric before generation could also help the agent choose exercises that demonstrate mastery rather than only covering topics.
Agent Mentor Your product has strong potential, but I found a few key improvements that could make it even better. I'd love to share my feedback and suggestions—please contact me at [email protected].
Premium Products
New to Fazier?
Find your next favorite product or submit your own. Made by @FalakDigital.
Copyright ©2025. All Rights Reserved