pre-pr-check

Run CI-equivalent lint, Vally eval, and full-eval checks locally.

14|3|Updated Mar 13, 2026
One-click install
npx skills add https://github.com/patrikborosch/AnalyticsPlatformAgents --skill pre-pr-check-patrikborosch
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pre-pr-check
Source: https://github.com/patrikborosch/AnalyticsPlatformAgents/tree/main/creator/.github/skills/pre-pr-check
Command: npx skills add https://github.com/patrikborosch/AnalyticsPlatformAgents --skill pre-pr-check-patrikborosch

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Contributors to the AnalyticsPlatformAgents repository frequently waste time waiting for CI results and fixing avoidable issues (structural lint errors, broken skill references, failing evals) that could have been caught locally before pushing a PR.

Core Features & Use Cases

  • Local CI Parity: Runs the exact three checks CI executes (quality structural lint, Vally eval for changed skills, filtered full-eval for multi-step changes) to ensure PRs pass on first push.
  • Guided Workflow: Walks users through prerequisites (ephemeral tenant login, workspace provisioning, capacity selection) and step-by-step execution for each check, including error handling and cleanup.
  • Use Case: A contributor modifying a Fabric skill can run this skill to validate their changes locally, catch issues in minutes instead of waiting 10+ minutes for CI, and avoid review rounds for failed checks.

Quick Start

Use the pre-pr-check skill to run all required local verification steps for your PR that touches skills, agents, common, or tests before you push your changes.

Frequently Asked Questions about pre-pr-check

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify pull request changes locally to pass CI on the first push?

To verify pull request changes locally, run structural linting, Vally evaluation for changed skills, and filtered full-eval runs to match CI checks before submission. This prevents failed CI runs by catching structural and behavioral issues early.

What is the best way to run local CI parity checks for skill definitions and agent configurations?

Local CI parity checks involve running the exact three checks CI executes: quality structural lint, Vally eval for changed skills, and filtered full-eval for multi-step changes. This ensures your modifications pass CI on the first attempt.

Why does my CI run fail after pushing PR changes to the AnalyticsPlatformAgents repository?

CI runs often fail due to avoidable structural lint errors, broken skill references, or failing Vally evals. Running local pre-submit checks catches these issues in minutes before pushing, eliminating the wait for CI feedback and subsequent rework.

Do I need an ephemeral tenant login to validate PR checks locally?

Yes, ephemeral tenant login and workspace provisioning are required prerequisites. The guided workflow walks you through capacity selection and step-by-step execution for each check, including MCP server configuration validation and post-run resource cleanup.

When do I need to run a filtered full-eval for multi-step behavior changes?

A filtered full-eval is required when PR changes involve multi-step behavior modifications. This local validation ensures that complex agent interactions and shared common code perform correctly before submission, avoiding failed CI runs and review rounds.