pr-tests

Executes the test plan checklist in a pull request body and posts results as PR comments.

1|Updated Apr 21, 2026
One-click install
npx skills add https://github.com/devbasex/ai-plugins --skill pr-tests-devbasex
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pr-tests
Source: https://github.com/devbasex/ai-plugins/tree/main/plugins/ndf/skills/pr-tests
Command: npx skills add https://github.com/devbasex/ai-plugins --skill pr-tests-devbasex

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manually running every item in a PR's test plan and reporting results is repetitive and error-prone. This Skill reads the Test Plan checklist from a pull request body, executes each item automatically, and records verifiable evidence directly on the PR. ## Core Features & Use Cases - Automated Test Plan Execution: Parses - [ ] checklist items from the PR body, classifies them as automatic or manual, and runs them in order. - Evidence-Based Reporting: Records the command, exit code, execution time, and scope for every item, then updates checkboxes or posts failure comments with error details. - Scoped vs Full Test Distinction: Clearly separates limited verification (e.g., pytest -k orders) from full test suite runs, and never marks unexecuted items as passed. - Use Case: After creating a PR with /ndf:pr, run this Skill to execute its test plan, check off passing items in the PR body, and post a summary comment with success/failure counts and success rate. ## Quick Start Ask the AI to run the test plan for PR number 42 and post the results as a comment on that pull request.

Frequently Asked Questions about pr-tests

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

FAQPage Schema
How do I run a pull request test plan automatically?▼

Provide the PR number, or let the Skill detect it from the current branch. It parses the Test Plan checklist from the PR body, executes each item in order, and updates checkboxes or posts comments with the results.

How are pull request test results reported on GitHub?▼

Passing items have their `- [ ]` checkbox updated to `- [x]` in the PR body. Failures and the final summary (total, passed, failed, skipped counts and success rate) are posted as PR comments with error details.

Can this Skill fix code when a PR test fails?▼

No. The Skill is constrained to test execution and result reporting only. It does not modify code, edit files, or perform git operations; failures are reported as PR comments with problems and suggested countermeasures.

What evidence is recorded for each executed test item?▼

Each item records four pieces of evidence: the executed command, exit code, execution timestamp, and target scope (filter condition or full run). Items missing any of these are not counted as executed.

What happens when a test plan item cannot be executed?▼

Unexecuted items are never marked as passed. The Skill reports them as not executed with the reason, and manual verification items note who checked what, since they have no command or exit code.