npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

gitlab-mr-ai

v1.2.0

Published

AI-powered GitLab Merge Request summarizer CLI tool

Readme

gitlab-mr-ai

npm version build status

AI-powered CLI tool to generate GitLab Merge Request summaries using Google Gemini.

Generate clear, structured summaries for your GitLab merge requests in a CI/CD-friendly way. This CLI analyzes MR diffs and uses Gemini to produce a concise description and key file changes.


Table of Contents


Installation

npm install -g gitlab-mr-ai
# or use npx without install
npx gitlab-mr-ai generate --mr 123

Usage (Local)

mr-ai generate --mr <merge-request-id> [options]

Example:

mr-ai generate --mr 927 --output console

To load environment variables locally, you can:

# using dotenv-cli (recommended for simplicity)
npm install -g dotenv-cli
dotenv -e .env -- mr-ai generate --mr 927

# or with native bash
source .env && mr-ai generate --mr 927

Example .env:

GEMINI_API_KEY=your-gemini-key
GITLAB_TOKEN=your-gitlab-token
GITLAB_PROJECT_ID=123
MR_AI_TICKET_PREFIXES=PL,OPS,WEB

Usage (GitLab CI/CD)

You can configure GitLab CI/CD to run this tool automatically or manually on merge requests.

Example .gitlab-ci.yml job:

mr-summary:
  stage: quality-gate
  image: node:20
  script:
    - export MR_AI_TICKET_PREFIXES="PL,OPS,WEB"
    - npx gitlab-mr-ai generate --mr "$CI_MERGE_REQUEST_IID" --output post
  rules:
    - if: $CI_PIPELINE_SOURCE == "merge_request_event"
      when: manual

MR_AI_TICKET_PREFIXES is optional. If omitted, the CLI falls back to PL.


CLI Parameters

Below is a list of CLI flags supported by mr-ai, including what they do and when to use them.

| Flag | Alias | Required | Description | |---|---|---|---| | --mr | -m | Yes | Merge request IID to summarize | | --template | -t | No | Template name (standard, bug-fix, general) or file path to .md | | --prompt | -p | No | Custom prompt file path (.txt) for Gemini | | --output | -o | No | console, file, or post (default: console) |

Flag Descriptions

  • --mr, -m (required) The Merge Request IID (internal ID visible in the GitLab URL, like /merge_requests/123). This is used to fetch the diff and identify which MR to summarize.

  • --template, -t Optional. Specifies which Markdown template to use. You can:

    • Pass a built-in template name: standard, bug-fix, or general
    • Or pass a file path to a custom template, for example ./my-template.md
  • --prompt, -p Optional. Specify a custom prompt file (.txt) to send to Gemini. If not provided, a default prompt will be used.

  • --output, -o Optional. Defines where the generated summary goes:

    • console (default): print to terminal
    • file: write to mr-summary-<mrId>.md
    • post: update the MR's description directly in GitLab

Examples:

# Basic usage (console output)
mr-ai generate --mr 1010

# Use general template explicitly
mr-ai generate --mr 1010 --template general

# Use external template and prompt, then write to file
mr-ai generate --mr 1010 --template ./my-template.md --prompt ./my-prompt.txt --output file

# Post directly to GitLab MR
mr-ai generate --mr 1010 --output post

Branch & Ticket Detection

If no --template is provided, the CLI tries to auto-detect based on your branch name:

  • fix/123-title -> uses bug-fix.md
  • feat/PL-456-feature -> uses standard.md
  • feat/OPS-456-feature -> uses standard.md when MR_AI_TICKET_PREFIXES includes OPS
  • otherwise -> fallback to general.md

Ticket ID is also auto-detected and injected into the template:

  • #123 for numeric branches
  • PL-456, OPS-456, or other configured prefixes if allowed by MR_AI_TICKET_PREFIXES

Prefix detection rules:

  • MR_AI_TICKET_PREFIXES accepts comma-separated values, for example PL,OPS,WEB
  • Prefix matching is normalized internally, so pl, ops is treated as PL,OPS
  • If MR_AI_TICKET_PREFIXES is empty, the default prefix is PL
  • Numeric tickets such as fix/123-login remain valid without any prefix config

General Usage & Fallbacks

You do not need to follow any branch naming convention.

To disable smart detection and always use the general fallback:

mr-ai generate --mr 9001 --template general

If you prefer to keep smart detection off by default, you can:

  • Use a generic or unmatched branch name, for example feature/login-screen
  • Pass --template general explicitly in every usage

Fallback behavior:

  • Use the general.md template if the branch format is not recognized
  • Skip ticket-specific detection if no valid ticket is found
  • Still render valid output from Gemini

You can also:

  • Provide your own Markdown template with --template ./custom.md
  • Provide your own prompt with --prompt ./custom.txt

Recommended per-project setup:

  • Set MR_AI_TICKET_PREFIXES in each repository's .env
  • Or define it in each GitLab project's CI/CD Variables
  • This allows different GitLab projects in the same group to use different ticket prefixes without code changes

Environment Variables

When running in GitLab CI/CD, configure these variables under GitLab Settings -> CI/CD -> Variables.

Set via .env or CI variables:

| Key | Required | Description | |---|---|---| | GEMINI_API_KEY | Yes | Google Generative AI API key | | GITLAB_TOKEN | Yes | GitLab personal access token with project read/write access | | GITLAB_PROJECT_ID | Yes | GitLab project numeric ID | | GITLAB_API_URL | No | GitLab API URL (default: https://gitlab.com/api/v4) | | GEMINI_MODEL | No | Gemini model ID (default: gemini-2.5-flash) | | MR_AI_GENAI_API_VERSION | No | Gemini API version override, for example v1 for stable endpoints | | MR_AI_TICKET_PREFIXES | No | Comma-separated ticket prefixes for branch detection (default: PL) |


Troubleshooting

Missing required env variable:

Missing required .env variables: GEMINI_API_KEY, GITLAB_TOKEN

Check your .env file or CI settings.

Template not found:

Failed to read template at ./custom.md

Make sure the file path exists and is valid.


License

MIT