- Home
- Skills
- DevOps & Cloud
- Senior Dev Skills: Meet Ava, Your Private Coding Mentor
More screenshots
Works with the AI Agents you already use
Senior Dev Skills: Meet Ava, Your Private Coding Mentor
Stop copying AI boilerplate blindly. Turn Claude into a private coding mentor who tests your logic, audits flaws, and teaches you to build it right.
Free
Senior Dev Skills: Meet Ava, Your Private Coding Mentor
Example session with this skill installed
I need to set up an S3 bucket with Terraform. Can you show me how to do it securely?
- Read your context and instructions
- Compiled the senior dev
I have initialized your study_vault/ and added a new reference file for S3 security. Instead of giving you the final code, we will follow the Socratic protocol.
Socratic Mentoring: S3 Security
I've updated study_vault/references/s3_security_invariants.md with the architectural mental model for object storage. Below is a baseline Terraform snippet.
Warning: This code contains intentional security flaws.
resource "aws_s3_bucket" "data_store" {
bucket = "company-sensitive-data"
tags = {
Environment = "Prod"
}
}
resource "aws_s3_bucket_public_access_block" "example" {
bucket = aws_s3_bucket.data_store.id
block_public_acls = false
block_public_policy = false
ignore_public_acls = false
restrict_public_buckets = false
}
The Micro-Challenge
Look at the aws_s3_bucket_public_access_block resource above.
- What specific risk does this configuration introduce to "company-sensitive-data"?
- Two critical security features are missing from the
aws_s3_bucketresource itself: Server-Side Encryption (SSE) and Versioning.
Task: Rewrite the snippet to block all public access and enable default AES256 encryption.
Next steps
- Review the security invariants in
study_vault/references/s3_security_invariants.md. - Provide the corrected Terraform HCL.
- Run
/quiz s3once you have patched the code to test your knowledge of S3 Bucket Policies.
Connects securely to your tools. The creator never sees your data.
About this skill
🚀 Why Ava Exists
Most AI coding assistants act as code vending machines: you ask for infrastructure, and the LLM spits out 50 lines of boilerplate. You hit copy-paste, deploy it, and understand none of the security risks, trade-offs, or underlying mechanics.
The result? AI dependency without skill acquisition.
Ava changes the paradigm. Instead of replacing your thinking, she acts as a senior technical lead sitting beside you in the terminal.
⚡ Key Capabilities
1. 🛡️ The Anti-Copilot Protocol
Ava refuses to give complete copy-paste answers. She follows a strict 5-stage scaffolding ladder: explaining mental models, drawing architectural diagrams, and presenting snippets with intentional security anti-patterns (over-permissive IAM, public CIDRs, root containers). She challenges you to identify the flaws and write the secure fixes yourself.
2. 📚 Autonomous Local Knowledge Vault
On launch, Ava automatically initializes a persistent study_vault/ inside your repository:
references/: Official documentation, architecture invariants, and cheat sheets.takeaways/: Chronological session logs, commands learned, and mental models.quizzes/: Active-recall question banks ready for Anki and Google NotebookLM.
3. 🌐 Autonomous Web Research & Curation
Studying a new cloud architecture or protocol? Ava uses your agent's web search tools to locate official, authoritative standards (RFCs, AWS, CIS Benchmarks, NIST), extracts core invariants, and commits formatted reference cheat sheets directly into your local vault.
4. 🎯 Dynamic Domain Customizability
While tuned out-of-the-box for DevSecOps & Cloud Security, you can point Ava at any technical discipline simply by saying:
"Ava, set target domain to: Kubernetes SRE"
(or Go/Rust Backend Systems, Offensive Security, Machine Learning Engineering)
📊 Feature Comparison: Generic AI Prompts vs. Ava
| Feature | Generic "Tutor" Prompts | Ava Socratic Skill |
|---|---|---|
| Code Generation | Vomits 50 lines of boilerplate | Socratic micro-challenges & flawed snippets |
| Workspace Memory | Forgets everything when session ends | Maintains persistent study_vault/ in repository |
| Security Auditing | Defaults to insecure boilerplate | Audits against CIS & NIST benchmarks |
| Documentation | Hallucinates outdated tutorials | Autonomously pulls & synthesizes live specs |
| Spaced Repetition | None | Auto-exports Anki & NotebookLM quiz decks |
🛠️ Quickstart for Users
For Claude Code / Terminal Agents:
Place the skill directory in your repository:
cp -r skills/ava-socratic-mentor .agents/skills/
Activate in chat:
"Ava, let's start a session on AWS IAM Identity Federation."
For Cursor / VS Code:
Add skills/ava-socratic-mentor/SKILL.md to your Cursor .cursorrules or Agent Prompts directory.
For Agensi / Claude Web:
Once installed, simply start your session in chat:
"Ava, let's start a session on AWS IAM Identity Federation."
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for DevOps & Cloud.
Requires local file system access for study_vault management. Requires web search capabilities for reference synthesis. Optimized for DevSecOps and Cloud domains.
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 1 day ago
- Free with an Agensi account