- Home
- Skills
- Prompt & Skill Engineering
- guided learning
guided learning
An adaptive technical tutor that uses structured session archiving and evidence-based loops to ensure true mastery.
Free
Works with the AI Agents you already use
guided learning
Example session with this skill installed
I want to learn how to implement JWT authentication in a FastAPI application. Let's start a guided session. I have basic Python knowledge but haven't used FastAPI security utilities.
- Read your context and instructions
- Compiled the guided learning
I have initialized a new guided learning session for JWT authentication in FastAPI.
Session UID: sess-550e8400-e29b-41d4-a716-446655440000
Archive Path: ./guided-sessions/sess-550e8400/
I have created the root macro scope for this session. Before we dive into implementation, I need to establish your baseline for how HTTP headers and token-based auth interact.
Initial Diagnostic Task
Explain what happens to a JWT on the client side after the server sends it back during a successful login, and how the client should include that JWT in a subsequent request to a protected endpoint.
| Node | Status | Goal |
|---|---|---|
jwt-header-exchange | unseen | Explain token transit and header structure |
fastapi-security-dep | unseen | Implement OAuth2PasswordBearer |
payload-validation | unseen | Decode and verify claims |
Next steps
- Provide your explanation of the client-side JWT flow.
- Once verified, we will map the specific FastAPI dependencies needed to automate this in your code.
Connects securely to your tools. The creator never sees your data.
About this skill
Software developers and technical educators often face the 'passive learning' problem where users skim documentation or code without actually retaining the logic. This skill replaces static study packs with an adaptive, evidence-based tutoring system. It enforces a strict explain-practice-evidence loop, ensuring learners demonstrate specific capabilities before moving to more complex nodes.
What it does
- Persistent Session Archiving tracks learner progress, evidence, and maps across different models or agent handoffs using a structured filesystem schema.
- Granular Scope Management isolates specific teaching topics into hierarchical chats, preventing context pollution and maintaining focus.
- Adaptive Diagnostic Loops identify exact knowledge gaps and route learners to prerequisites or advanced nodes based on observable output.
- Evidence-First Validation requires learners to perform calculations, critiques, or code modifications rather than just confirming they understand.
- Dynamic Resource Selection triggers specific modalities like Mermaid diagrams or code traces only when they directly enable the next learner action.
How it works
- Initialize Session creates a durable archive with a unique Session UID and directory structure for state persistence.
- Orient and Map establishes the target capability and places the learner on a provisional path based on an initial diagnostic task.
- Elicit Evidence requests the smallest informative output—a prediction, code change, or critique—to verify node security.
- Update and Persist records demonstrated evidence and transitions to the next node in the
SESSION.mdandCHECKPOINT.mdfiles.
Frameworks & tools
The skill is framework-agnostic but optimized for technical domains involving code, mathematics, and systems design. It uses Mermaid for architectural visualization and standard filesystem operations for session persistence.
Why this beats prompting it yourself
Standard prompts often dump information or provide generic quizzes that don't prove mastery. This skill implements a sophisticated state machine that prevents "node hopping" and ensures every step is justified by learner evidence stored in a portable archive.
Use cases
- Technical Onboarding helping new hires master a specific internal codebase or architecture through active modification tasks.
- Interview Preparation simulating technical assessments with real-time feedback and gap analysis.
- Troubleshooting Mentorship guiding a developer through a complex bug by eliciting diagnostic hypotheses rather than giving the fix.
Known limitations
Requires a writable filesystem path for full session persistence. It does not support real-time collaborative writing by multiple agents to the same scope simultaneously.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Prompt & Skill Engineering.
Requires filesystem access for session archiving. One active writer per scope ID. Best for technical domains; not a substitute for real-time human clinical instruction.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 14 days ago
- Free with an Agensi account
- Software
Git