managing-deployment-rollbacks
About
This skill automates deployment rollbacks when triggered by phrases like "rollback deployment" or "revert to previous version," typically used after a failed deployment. It executes the rollback process with built-in safety checks and provides options for verification. This ensures a reliable and controlled recovery to a previous stable state.
Quick Install
Claude Code
Recommended/plugin add https://github.com/jeremylongshore/claude-code-plugins-plusgit clone https://github.com/jeremylongshore/claude-code-plugins-plus.git ~/.claude/skills/managing-deployment-rollbacksCopy and paste this command in Claude Code to install this skill
Documentation
Overview
This skill enables Claude to manage and execute deployment rollbacks, ensuring a stable and reliable system. It automates the process of reverting to a previous, known-good deployment state, minimizing downtime and potential disruptions.
How It Works
- Identifying the Issue: Claude analyzes the user's request to confirm the need for a rollback due to a failed deployment.
- Executing Rollback: Claude executes the rollback command, reverting the deployment to the previously known stable version.
- Verification: Claude verifies the success of the rollback by checking the application's status, health checks, and logs.
When to Use This Skill
This skill activates when you need to:
- Recover from a failed deployment.
- Revert to a previous version of an application.
- Address issues caused by a recent deployment.
Examples
Example 1: Rolling back a faulty deployment
User request: "Rollback deployment to the previous stable version due to errors."
The skill will:
- Execute the
rollback-deploycommand. - Confirm the successful rollback to the previous version.
Example 2: Reverting after performance degradation
User request: "Revert to the last known good deployment. Performance has degraded significantly since the last update."
The skill will:
- Execute the
rollback-deploycommand. - Verify the application's performance returns to normal levels.
Best Practices
- Verification: Always verify the success of the rollback by checking application health and performance.
- Monitoring: Implement continuous monitoring to detect deployment issues early.
- Documentation: Maintain clear documentation of deployment versions and rollback procedures.
Integration
This skill can be used in conjunction with other monitoring and alerting tools to automatically trigger rollbacks when issues are detected. It also integrates with deployment pipelines for automated rollback execution.
GitHub Repository
Related Skills
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
evaluating-llms-harness
TestingThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
llamaguard
OtherLlamaGuard is Meta's 7-8B parameter model for moderating LLM inputs and outputs across six safety categories like violence and hate speech. It offers 94-95% accuracy and can be deployed using vLLM, Hugging Face, or Amazon SageMaker. Use this skill to easily integrate content filtering and safety guardrails into your AI applications.
langchain
MetaLangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.
