Skip to content

intent.md

Latest commit

 

History

History
116 lines (83 loc) · 2.2 KB

intent.md

File metadata and controls

116 lines (83 loc) · 2.2 KB

FeaturePulse – Pull Request Intent Policy

Purpose

This document defines the intended purpose and acceptable behavior of pull requests for this repository. FeaturePulse uses this file as the source of truth when evaluating whether a pull request aligns with project goals.

What Is “Intent”?

Intent describes why a change exists, not just what it changes.

FeaturePulse infers intent from:

  • Pull request title
  • Pull request description
  • Commit context
  • Files modified

Supported Intent Categories

1. Documentation Update

Changes that improve or modify:

  • README files
  • Markdown documentation
  • Code comments

Risk Level: Low
Default Decision: APPROVE

2. Bug Fix

Changes that correct:

  • Broken functionality
  • Incorrect logic
  • Runtime errors

Risk Level: Medium
Default Decision: APPROVE
Note: Must not introduce unrelated changes.

3. New Feature

Changes that introduce:

  • New functionality
  • New modules or endpoints
  • New user-facing behavior

Risk Level: High
Default Decision: WARN
Note: Should include clear description and intent.

4. Refactor

Changes that:

  • Improve code structure
  • Improve readability or maintainability
  • Do not change behavior

Risk Level: Medium
Default Decision: APPROVE

5. Test Improvement

Changes that:

  • Add tests
  • Improve test coverage
  • Fix failing tests

Risk Level: Low
Default Decision: APPROVE

6. General Improvement

Minor changes that do not fit other categories.

Risk Level: Low
Default Decision: APPROVE

Branch-Specific Rules

Master Branch

Pull requests targeting master must:

  • Clearly match one of the supported intent categories
  • Avoid experimental or unfinished features
  • Align with production stability

FeaturePulse applies stricter evaluation for master merges.

How FeaturePulse Uses This File

FeaturePulse:

  1. Reads this intent.md file
  2. Compares PR details against defined intent rules
  3. Uses AI reasoning to determine:
    • Detected intent
    • Risk level
    • Merge decision (APPROVE / WARN / BLOCK)
  4. Posts results as:
    • A GitHub Check
    • A pull request comment

Example Output