KS
Killer-Skills

ship — how to use ship how to use ship, ship setup guide, ship alternative, ship vs eval frameworks, ship install, what is ship, ship for developers, ship eval framework, ship version management

v1.0.0
GitHub

About this Skill

Perfect for DevOps Agents needing streamlined testing and version management on the dev branch ship is a code-first eval framework that enables efficient testing and versioning through a series of commands and file updates, including changelog management.

Features

Runs tests in parallel using -n auto for efficient evaluation
Manages version bumps based on existing versions in the changelog
Updates docs/changelog.mdx to reflect new version releases
Verifies the current branch to ensure development on the dev branch
Supports stopping and reporting test failures for quality assurance
Automates changelog updates with the current date and version

# Core Topics

camronh camronh
[0]
[0]
Updated: 3/6/2026

Quality Score

Top 5%
36
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
Cursor IDE Windsurf IDE VS Code IDE
> npx killer-skills add camronh/EZVals/ship

Agent Capability Analysis

The ship MCP Server by camronh is an open-source Categories.community integration for Claude and other AI agents, enabling seamless task automation and capability expansion. Optimized for how to use ship, ship setup guide, ship alternative.

Ideal Agent Persona

Perfect for DevOps Agents needing streamlined testing and version management on the dev branch

Core Value

Empowers agents to automate testing workflows using parallel execution with `-n auto`, manage version bumps based on changelog updates, and maintain accurate release notes in `docs/changelog.mdx`

Capabilities Granted for ship MCP Server

Automating test suites on the dev branch
Generating version bumps based on changelog updates
Updating release notes in `docs/changelog.mdx`

! Prerequisites & Limits

  • Requires presence on the `dev` branch
  • Dependent on successful test execution
Project
SKILL.md
1.9 KB
.cursorrules
1.2 KB
package.json
240 B
Ready
UTF-8

# Tags

[No tags]
SKILL.md
Readonly

Follow these steps in order:

1. Confirm on Dev

  • Verify we're on the dev branch
  • If not, stop and ask

2. Run Tests

  • Run tests and ensure all tests pass. (Use -n auto to run tests in parallel)
  • If tests fail, STOP and report the failures

3. Version Bump

  • Read docs/changelog.mdx to see what's in the ## Unreleased section
  • Suggest the next version (based on existing versions in the changelog)
  • Update docs/changelog.mdx: change ## Unreleased to ## <version> - <today's date>
  • Update pyproject.toml version to match
  • Run uv lock to sync the lock file with the new version
  • Summarize the changes for this version in a human friendly way. Focus on impactful changes and user facing changes.

4. Update Skill Version

  • Update the version comment in ezvals/skills/evals/SKILL.md (the <!-- Version: X.X.X --> line) to match the release version
  • The skill version MUST match the release tag — the Sync Skill to Marketplace workflow will fail otherwise

5. Commit & Push Dev

bash
1git add -A 2git commit -m "release <version>" 3git push origin dev

6. Merge to Main

bash
1git checkout main 2git pull origin main 3git merge dev 4git push origin main

7. Tag & Push

bash
1git tag v<version> 2git push origin v<version>

8. Return to Dev

bash
1git checkout dev

PyPI Publishing (Automatic)

Publishing to PyPI is handled automatically by GitHub Actions:

  • Dev builds (.github/workflows/publish-dev.yml): Every push to main publishes a dev version (0.0.0.dev{timestamp}) to PyPI
  • Release builds (.github/workflows/publish.yml): Pushing a tag like v0.1.0 triggers a release publish to PyPI

You don't need to manually run uv publish - just push the tag and the workflow handles it.

Error Handling

  • If tests fail → stop and report
  • If merge conflicts occur → stop and ask for help
  • If not on dev → stop and ask

Related Skills

Looking for an alternative to ship or building a Categories.community AI Agent? Explore these related open-source MCP Servers.

View All

widget-generator

Logo of f
f

widget-generator is an open-source AI agent skill for creating widget plugins that are injected into prompt feeds on prompts.chat. It supports two rendering modes: standard prompt widgets using default PromptCard styling and custom render widgets built as full React components.

149.6k
0
Design

chat-sdk

Logo of lobehub
lobehub

chat-sdk is a unified TypeScript SDK for building chat bots across multiple platforms, providing a single interface for deploying bot logic.

73.0k
0
Communication

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication