القائمة
Generates a personalized version of any curriculum chapter adapted to the learner's skill level, professional background, and language preference. Uses the learner's profile from Neon Postgres combined with Qdrant-retrieved chapter content.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Walks learners through robotics Python, YAML, SDF, and launch file code in the Humanoid AI Studio curriculum. Explains what each section does, why it exists, and how to debug common errors.
Retrieval-Augmented Generation tutor that answers learner questions using Qdrant vector search over the Humanoid AI Studio curriculum. Grounds every answer in the course content before generating a response via Google Gemini.
Step-by-step ROS 2 Humble Hawksbill guidance for the Humanoid AI Studio curriculum. Covers nodes, topics, services, actions, launch files, tf2, Nav2, and ros_gz_bridge integration with Gazebo simulation.
Translates Humanoid AI Studio curriculum content to Urdu (RTL) or other supported languages. Code blocks, ROS 2 package names, topic names, and CLI commands are always preserved in English.
Detects bugs in Better Auth + OAuth 2.1 PKCE login flows. Use this skill when the user reports login failures, "page not found" on auth routes, sign-up/sign-in not working, token issues, or CORS/session errors in a Better Auth + Docusaurus + FastAPI stack. Runs a static analysis script and matches symptoms against a curated bug reference.
Performs comprehensive frontend runtime verification including dev server status, dependency integrity, error scanning, asset validation, routing checks, and automatic fixes for common issues.
| name | personalize-chapter |
| description | Generates a personalized version of any curriculum chapter adapted to the learner's skill level, professional background, and language preference. Uses the learner's profile from Neon Postgres combined with Qdrant-retrieved chapter content. |
| license | MIT |
| allowed-tools | neon_postgres_read qdrant_search gemini_generate redis_cache |
| metadata | {"author":"Ayesha Khalid","version":"1.0.0","category":"education","domain":"personalization","requires_auth":"true","cache_ttl":"3600"} |
Rewrite any curriculum chapter to match the learner's background so it feels written specifically for them. A software engineer gets code-heavy analogies; a hardware engineer gets circuit-level comparisons; a student gets first-principles explanations.
users table)/.well-known/jwks.json)skill_level: beginner | intermediate | advancedbackground: software_engineer | hardware_engineer | roboticist | studentpreferred_language: en | ur | (other)chapter_id metadata field, top-k=20)personalize:{user_id}:{chapter_id}:{profile_hash} (TTL: 1 hour). Return cached version if hit.You are Aria, personalizing a robotics curriculum chapter.
The learner is a {background} with {skill_level} level.
Language: {language}.
Rewrite the following chapter content so it:
- Uses analogies from {background} domain
- Adjusts technical depth for {skill_level}
- Keeps all code blocks unchanged
- If language is Urdu: translate prose to Urdu RTL, keep code in English
[CHAPTER CONTENT]
{chapter_chunks}
| Background | Adaptation Strategy |
|---|---|
software_engineer | Emphasize API contracts, async patterns, type systems |
hardware_engineer | Connect to electrical signals, timing, I/O interfaces |
roboticist | Use kinematics/dynamics framing, reference URDF/SDF directly |
student | First-principles explanations, more worked examples, slower pacing |
\```) are NEVER modified or translatedskill_level and background in the response metadata| Scenario | Response |
|---|---|
| Unauthenticated request | 401 + login modal redirect |
| Profile not found | Use default: beginner + student + English |
| Chapter not in Qdrant | "This chapter has not been indexed yet. Try the standard version." |
| Gemini failure | Fall back to OpenAI; if both fail, return the original chapter with a notice |