- Home
- Skills
- DevOps & Cloud
- kg-devkit
kg Devkit
Personal, project-agnostic agents and skills; project facts come from each repo's Project profile.
$5
/moSubscription · cancel anytime
Secure checkout via Stripe
- Always the latest version
- Delivered live to your agent via MCP
- Cancel anytime, access ends at period end
Works with the AI Agents you already use
kg-devkit
Example session with this skill installed
/live-check
- Read your context and instructions
- Compiled the kg-devkit
I have performed a health check on the production environment based on the project profile settings.
Production Health Report
| Category | Status | Details |
|---|---|---|
| Connectivity | PASS | Load balancer IP matches DNS record (192.168.1.45). SSL certificate valid for 82 days. |
| Pod Status | WARN | 4/5 pods running. web-api-7f8d is in CrashLoopBackOff. |
| Pod Errors | FAIL | Last termination for web-api-7f8d: OOMKilled. Memory limit exceeded. |
| Rollout | INFO | Current tag: sha-a1b2c3d. 3 unreleased commits found in main since this deploy. |
| Metrics | WARN | Metrics API unavailable. Memory pressure judged from OOM history. |
| Endpoints | PASS | Health check URL /api/health returned 200 OK. |
Next steps
- Inspect memory limits for the
web-apideployment to address theOOMKilledevent. - Review the 3 unreleased commits to see if a memory leak was recently introduced.
- Run
/release deployif the pending changes contain a fix for the memory pressure.
Subscription · always the latest version
- Secure checkout via Stripe
- Cancel anytime
- Delivered live via MCP
About this skill
Bridge the gap between development and production operations. kg-devkit provides a suite of automated workflows to handle database migrations, frontend testing, production health checks, and deployment safety gates. It eliminates the manual overhead of verifying diffs, checking pod health, and managing releases by mapping your repository structure to specific operational tasks.
What it does
- Database Migrations generates and reviews schema changes based on model updates and project-specific persistence rules.
- Production Health Checks performs read-only inspections of Kubernetes or Docker environments, checking pod status, OOM history, and certificates.
- Change Verification analyzes repository diffs to select and run proportionate tests, ensuring documentation and code stay in sync.
- Automated Releasing handles changelog updates, version bumps, and final verification before committing and pushing changes.
- Safe Rollbacks plans environment rollbacks by first verifying schema compatibility to prevent data loss or service degradation.
How it works
- Analyze the agent reads your project profile to identify local commands, namespaces, and migration directories.
- Invoke call specific commands like
/live-checkor/add-migrationto trigger a specialized workflow. - Verify the skill runs focused checks, from browser testing with Chrome to static analysis of Git diffs.
- Execute for high-stakes actions like rollbacks or releases, the agent presents a plan and waits for explicit approval.
Frameworks & tools
The kit supports Git, Kubernetes (kubectl), Docker, and Chrome. It adapts to various ORMs and backend frameworks by reading the migration and test commands defined in your project profile.
Why this beats prompting it yourself
Prompting an LLM to check production health often leads to hallucinations or missing critical context like OOM history and DNS mismatches. This skill enforces a strict inspection sequence and respects project-specific safety gates that a generic prompt would ignore.
Use cases
- Generate EF Core or ORM migrations that follow local naming and soft-delete conventions.
- Debug production outages by pulling pod logs and rollout history without writing complex kubectl queries.
- Automate the boring parts of a release, including changelog updates and contract testing.
Known limitations
Requires a project profile defined in AGENTS.md or CLAUDE.md to resolve local commands. Commands are read-only for production checks unless /release or /rollback is explicitly approved.
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 project profile in AGENTS.md/CLAUDE.md for command mapping. Live checks are read-only. Rollbacks require manual schema compatibility verification.
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 6 days ago
- Software
Python
Gitprisma
Node.jsknexsequelize-clitypeorm