---
title: "starlight-academy"
description: "Practice with Academy teaching guides, source-backed challenges and useful AI feedback in your own conversation."
canonical: https://agentpluginsdirectory.com/plugins/starlight-academy
last-updated: 2026-09-13
---

# starlight-academy
Practice with Academy teaching guides, source-backed challenges and useful AI feedback in your own conversation.
- Slug: starlight-academy
- Publisher: Starlight Intelligence Systems
- Repository: https://github.com/frankxai/starlight-academy-plugin
- Manifest: plugins/starlight-academy/plugin.json
- Version: 0.1.0
- License: Apache-2.0
- Category (editorial): other
- Skills: 1 (learn-with-starlight)
- MCP servers: 1 (starlight-academy)
- Stars: 0
- Repository created: 2026-09-12
- Repository last pushed: 2026-09-12
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/starlight-academy
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What starlight-academy does, in the publisher's words

Public teaching material for your AI conversation: choose a challenge, make your own first attempt, practice with a guide, critique, revise and test transfer.

The plugin provides one guiding skill and connects four public MCP tools: get_academy, get_teacher, get_challenge, get_feedback_guide. The server is https://starlightintelligence.academy/mcp (Streamable HTTP). It accepts catalog IDs only. No drafts, accounts, private data, model calls, student database, automatic training, credentials or external actions.

Your chosen host provides the AI conversation under its own data policy. Teachers are fictional teaching methods, not separately running models or human instructors. Switching teacher roles is not independent review.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/frankxai/starlight-academy-plugin/HEAD/plugins/starlight-academy/README.md

## Skills

- learn-with-starlight: Learn with Starlight Academy when the user asks for its teachers, challenges, guided practice, or help giving useful feedback to AI. Choose a public mission, preserve a first attempt, coach with bounded hints, critique, revise and test transfer.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.

## MCP servers

- starlight-academy: transport: streamable-http; url: https://starlightintelligence.academy/mcp

Read from the plugin's own mcp.json. Environment variable names only, never values.