velocirepo 0.3.6

All your open-source and social-media metrics in one place

velocirepo

Track your project's pulse across package registries, GitHub, and the web.

Table of contents

• Overview
• Installation
• Configuration
• Usage
• Data storage
• Migrating data
• Querying the data
• Exporting data
• Using the data with other tools
• Indicators
• Generating badges
• Views
• MCP server

Overview

velocirepo collects metrics from multiple sources, stores them as JSONL files, and exposes them via SQL (powered by DuckDB). It's designed to run on a schedule (e.g., nightly via GitHub Actions) and commit the results to git, giving you a permanent record of your project's growth.

Supported sources:

Source	What it tracks
GitHub Events	Individual events (stars, forks, issues, PRs) with user and timestamp
GitHub Traffic	Daily page views and git clones (requires admin access)
PyPI	Daily download counts
CRAN	Daily download counts
Homebrew	Install counts (30-day, 90-day, 365-day, lifetime)
Plausible	Daily pageviews, visitors, visits
OpenVSX	Total downloads, reviews, ratings
YouTube	Views, likes, comments, subscribers (channel and per-video)

Installation

velocirepo can be used as a GitHub Action for automated nightly fetching, or installed locally for ad-hoc use.

GitHub Actions (recommended)

name: Fetch Metrics

on:
  schedule:
    - cron: '0 1 * * *'
  workflow_dispatch:

permissions:
  contents: write

jobs:
  fetch:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - uses: jeroenjanssens/velocirepo@v0
        with:
          github-token: ${{ secrets.GH_TOKEN }}
          # plausible-token: ${{ secrets.PLAUSIBLE_TOKEN }}
          # youtube-token: ${{ secrets.YOUTUBE_TOKEN }}

      - name: Commit and push
        run: |
          git config --local user.email "github-actions[bot]@users.noreply.github.com"
          git config --local user.name "github-actions[bot]"
          git add velocirepo/
          git diff --staged --quiet || git commit -m "Update metrics - $(date -u +'%Y-%m-%d')"
          git push

Or generate this workflow automatically (no installation required):

uvx velocirepo install-ci

Local installation

Homebrew (macOS / Linux)

brew install jeroenjanssens/tap/velocirepo

Scoop (Windows)

scoop bucket add jeroenjanssens https://github.com/jeroenjanssens/scoop-bucket
scoop install velocirepo

uv

uv tool install velocirepo

Or run without installing:

uvx velocirepo fetch

Go

go install github.com/jeroenjanssens/velocirepo/cmd/velocirepo@latest

From source

git clone https://github.com/jeroenjanssens/velocirepo.git
cd velocirepo
go build -o bin/velocirepo ./cmd/velocirepo
cp bin/velocirepo ~/.local/bin/

Shell installer (macOS / Linux)

curl -sSfL https://raw.githubusercontent.com/jeroenjanssens/velocirepo/main/install.sh | sh

This downloads the latest binary to ./bin/velocirepo. Set INSTALL_DIR to install elsewhere:

curl -sSfL https://raw.githubusercontent.com/jeroenjanssens/velocirepo/main/install.sh | INSTALL_DIR=/usr/local/bin sh

Download binary

Pre-built binaries for Linux, macOS, and Windows are available on the Releases page.

Shell completion

# Zsh (add to ~/.zshrc)
eval "$(velocirepo completion zsh)"

# Bash (add to ~/.bashrc)
eval "$(velocirepo completion bash)"

# Fish
velocirepo completion fish | source

Configuration

Create a velocirepo.toml in your project root:

[projects.my-project]
name = "My Project"
github-events = "owner/repo"
github-traffic = "owner/repo"
pypi = "my-package"

[projects.other-project]
name = "Other Project"
github-events = ["owner/other", "owner/other-utils"]
cran = "other"
homebrew = "other"
youtube = "@ChannelHandle"

Each source field accepts either a single string or an array of strings, so you can track multiple repositories or packages under one project.

The github-traffic source fetches daily page views and clone counts. GitHub only retains this data for 14 days, so velocirepo preserves it before it's lost. It requires a token with Administration:read permission (or the repo scope for classic tokens).

Or initialize one interactively (auto-detects sources from your repository):

velocirepo init

velocirepo looks for velocirepo.toml by walking up from the current directory. Override with --config or the VELOCIREPO_CONFIG environment variable.

Environment variables

Variable	Description
GITHUB_TOKEN	GitHub personal access token (increases rate limits)
PLAUSIBLE_TOKEN	Plausible API token
YOUTUBE_TOKEN	YouTube Data API key
VELOCIREPO_CONFIG	Path to config file

These can also be set in a .env file in the current directory.

Usage

velocirepo fetch                 Fetch from all configured sources
velocirepo fetch-github          Fetch GitHub events (stars, forks, issues, PRs)
velocirepo fetch-traffic         Fetch GitHub traffic (views and clones)
velocirepo fetch-pypi            Fetch PyPI download counts
velocirepo fetch-cran            Fetch CRAN download counts
velocirepo fetch-homebrew        Fetch Homebrew install counts
velocirepo fetch-plausible       Fetch Plausible analytics
velocirepo fetch-openvsx         Fetch Open VSX extension metrics
velocirepo fetch-youtube         Fetch YouTube metrics

velocirepo query <sql>           Run a SQL query against the metrics data
velocirepo schema                Show table schemas
velocirepo show-indicators       Print built-in indicator definitions as TOML

velocirepo export <dir>          Export data to Parquet or CSV files

velocirepo badge stars           Generate a stars badge
velocirepo badge forks           Generate a forks badge
velocirepo badge downloads       Generate a downloads badge
velocirepo badge pageviews       Generate a pageviews badge
velocirepo badge custom          Generate a badge from a custom query

velocirepo init                  Create a new velocirepo.toml
velocirepo add-project           Add a project to the config
velocirepo update-project        Update a project's configuration
velocirepo remove-project        Remove a project from the config
velocirepo rename-project        Rename a project's ID
velocirepo list-projects         List configured projects
velocirepo show-project          Show project details
velocirepo import-projects       Bulk-import from GitHub org/user or file
velocirepo validate-projects     Validate source URLs

velocirepo add-view <name>       Scaffold a new view
velocirepo remove-view <name>    Remove a view and its output
velocirepo list-views            List all views
velocirepo show-view <name>      Show details about a view
velocirepo render-view <name>    Render a single view
velocirepo render-views          Render all or filtered views
velocirepo serve-view <name>     Start a dev server for a view

velocirepo build-db              Build the DuckDB database file for external tools
velocirepo migrate               Migrate data to the latest schema version

velocirepo install-ci            Generate a GitHub Actions workflow
velocirepo sync-secrets          Sync .env secrets to GitHub Actions

velocirepo version               Print version information

Data storage

All data is stored as JSONL files at velocirepo/data/<source>/<project-id>/<date>.jsonl. This entire directory is meant to be committed to git — it's your permanent metric history.

Metrics sources

Sources like PyPI, CRAN, Homebrew, Plausible, OpenVSX, and GitHub Traffic store one JSON object per metric per day:

{"source":"pypi","metric":"daily_downloads","project_id":"plotnine","target":"plotnine","date":"2026-06-15","value":1523}
{"source":"openvsx","metric":"total_downloads","project_id":"quarto","target":"quarto/quarto","date":"2026-06-15","value":1250000}

Metric names are prefixed with daily_ (for deltas — values that reset each day) or total_ (for snapshots — cumulative totals at a point in time). Homebrew metrics use their own naming (downloads_30d, downloads_365d, etc.).

Fields:

Field	Description
source	Source name (pypi, cran, homebrew, plausible, openvsx, github-traffic)
metric	Metric name (e.g., daily_downloads, total_views, daily_pageviews)
project_id	Project ID from your config
target	Specific package, repo, site, or extension being tracked
date	Date of the measurement (YYYY-MM-DD)
value	Integer value
tags	Optional key-value metadata (e.g., {"video_id": "..."} for YouTube)

GitHub Events source

The GitHub Events source stores individual events rather than pre-computed counts, giving you full historical detail including who performed each action and when:

{"source":"github","event_type":"star","project_id":"quarto","github_repo":"quarto-dev/quarto-cli","datetime":"2026-06-15T14:23:01Z","user":"alice"}
{"source":"github","event_type":"fork","project_id":"quarto","github_repo":"quarto-dev/quarto-cli","datetime":"2026-06-15T09:11:44Z","user":"bob"}

Fields:

Field	Description
source	Always github
event_type	One of: star, fork, issue_open, issue_close, pr_open, pr_merge
project_id	Project ID from your config
github_repo	The owner/repo being tracked
datetime	Full timestamp of the event (ISO 8601)
user	GitHub username who performed the action

These events are automatically aggregated into daily counts in the metrics DuckDB view (as daily_stars, daily_forks, etc.) so you can query them alongside other sources.

YouTube index

The YouTube source also writes an index.jsonl file at velocirepo/data/youtube/<project-id>/index.jsonl containing video metadata:

{"video_id":"ML3q7Ok4hJg","title":"God-Tier Developer Roadmap","published_at":"2024-03-15T16:00:00Z","channel":"@Fireship","duration":423,"tags":["programming","roadmap"]}

This is exposed as the youtube_index DuckDB view, allowing you to join video titles and publish dates with metrics data.

Concatenation

Daily JSONL files are automatically concatenated into monthly and yearly files once a period is complete. For example, once all days in January 2026 have been fetched, they're concatenated into 2026-01.jsonl. This keeps the file count manageable for long-running histories. The original daily files are removed after concatenation.

Repository layout

You can either keep metrics in the same repository as your code, or create a dedicated metrics repository. A separate repo is useful when you want to track multiple projects in one place or keep metric history out of your main codebase.

Migrating data

When a new version of velocirepo changes the on-disk data format, it tracks this with a schema version number in velocirepo/data/.schema-version. Commands like fetch, query, and export will refuse to run against stale data:

Error: data schema is at version 0, but version 1 is required; run `velocirepo migrate` to update

To migrate your data to the latest schema:

velocirepo migrate

If you've copied in data from an older schema (e.g., merged data from another repository), some files may be at a different version than what .schema-version claims. Use --force to re-run all migrations from scratch:

velocirepo migrate --force

This is safe to run repeatedly — all migrations are idempotent.

Querying the data

The query command reads JSONL files directly using DuckDB and exposes five views:

View	Description
metrics	Unified time-series: all sources including aggregated GitHub events
indicators	Derived growth rate and trend for daily metrics (28-day windows)
github_events	Raw GitHub events with user and timestamp
youtube_index	Video metadata (title, publish date, channel, duration)
projects	Project metadata from your config

Daily star counts

velocirepo query "
  SELECT project, date, value AS stars
  FROM metrics
  WHERE source = 'github' AND metric = 'daily_stars'
  ORDER BY date DESC
  LIMIT 5
"

┌──────────────────┬────────────┬───────┐
│     project      │    date    │ stars │
├──────────────────┼────────────┼───────┤
│ ggsql            │ 2026-06-22 │ 1     │
│ gt               │ 2026-06-22 │ 1     │
│ images-workbench │ 2026-06-22 │ 1     │
│ rmarkdown        │ 2026-06-22 │ 1     │
│ pointblank       │ 2026-06-22 │ 1     │
└──────────────────┴────────────┴───────┘

Total stars per project

velocirepo query "
  SELECT p.name, SUM(value) AS stars
  FROM metrics m
  JOIN projects p ON m.project = p.id
  WHERE m.source = 'github' AND m.metric = 'daily_stars'
  GROUP BY p.name
  ORDER BY stars DESC
  LIMIT 5
"

┌─────────────┬───────┐
│    name     │ stars │
├─────────────┼───────┤
│ ggplot2     │ 6877  │
│ cheatsheets │ 6360  │
│ shiny       │ 5655  │
│ Shiny for R │ 5600  │
│ Quarto      │ 5274  │
└─────────────┴───────┘

Monthly star activity using raw events

The github_events view gives you access to individual events when you need per-user or per-timestamp detail:

velocirepo query "
  SELECT date_trunc('month', datetime)::DATE AS month, COUNT(*) AS stars
  FROM github_events
  WHERE project = 'quarto' AND event_type = 'star'
  GROUP BY month
  ORDER BY month DESC
  LIMIT 5
"

┌────────────┬───────┐
│   month    │ stars │
├────────────┼───────┤
│ 2026-02-01 │ 40    │
│ 2026-01-01 │ 107   │
│ 2025-12-01 │ 91    │
│ 2025-11-01 │ 97    │
│ 2025-10-01 │ 85    │
└────────────┴───────┘

Top YouTube videos by views

velocirepo query "
  SELECT yi.title, m.value AS views
  FROM metrics m
  JOIN youtube_index yi ON m.tags->>'video_id' = yi.video_id
  WHERE m.source = 'youtube' AND m.metric = 'total_views'
  ORDER BY m.value DESC
  LIMIT 5
"

Latest metrics across sources

velocirepo query "
  SELECT project, source, metric, date, value
  FROM metrics
  WHERE source != 'github'
  ORDER BY date DESC
  LIMIT 5
"

┌───────────────────┬─────────┬─────────────────┬────────────┬─────────┐
│      project      │ source  │     metric      │    date    │  value  │
├───────────────────┼─────────┼─────────────────┼────────────┼─────────┤
│ shiny-r           │ openvsx │ total_downloads │ 2026-06-22 │ 2404755 │
│ shiny-r           │ openvsx │ rating          │ 2026-06-22 │ 5       │
│ shiny-r           │ openvsx │ reviews         │ 2026-06-22 │ 2       │
│ pointblank-python │ pypi    │ daily_downloads │ 2026-06-22 │ 604     │
│ orbital-r         │ cran    │ daily_downloads │ 2026-06-22 │ 37      │
└───────────────────┴─────────┴─────────────────┴────────────┴─────────┘

Output formats

By default, results are printed as a table. Use --json, --csv, or --parquet for machine-readable output:

velocirepo query --csv "SELECT project, metric, value FROM metrics LIMIT 3"
velocirepo query --json "SELECT project, metric, value FROM metrics LIMIT 3"
velocirepo query --parquet "SELECT * FROM metrics" > metrics.parquet

The schema command shows all available columns:

velocirepo schema

┌───────────────┬──────────────┬───────────┐
│ TABLE         │ COLUMN       │ TYPE      │
├───────────────┼──────────────┼───────────┤
│ github_events │ project      │ VARCHAR   │
│ github_events │ source       │ VARCHAR   │
│ github_events │ event_type   │ VARCHAR   │
│ github_events │ github_repo  │ VARCHAR   │
│ github_events │ datetime     │ TIMESTAMP │
│ github_events │ user         │ VARCHAR   │
│ indicators    │ project      │ VARCHAR   │
│ indicators    │ source       │ VARCHAR   │
│ indicators    │ target       │ VARCHAR   │
│ indicators    │ metric       │ VARCHAR   │
│ indicators    │ indicator    │ VARCHAR   │
│ indicators    │ date         │ DATE      │
│ indicators    │ value        │ DOUBLE    │
│ indicators    │ tags         │ JSON      │
│ metrics       │ project      │ VARCHAR   │
│ metrics       │ source       │ VARCHAR   │
│ metrics       │ target       │ VARCHAR   │
│ metrics       │ metric       │ VARCHAR   │
│ metrics       │ date         │ DATE      │
│ metrics       │ value        │ BIGINT    │
│ metrics       │ tags         │ JSON      │
│ projects      │ id           │ VARCHAR   │
│ projects      │ name         │ VARCHAR   │
│ projects      │ description  │ VARCHAR   │
│ projects      │ color        │ VARCHAR   │
│ projects      │ tags         │ VARCHAR[] │
│ projects      │ website      │ VARCHAR   │
│ projects      │ logo         │ VARCHAR   │
│ youtube_index │ video_id     │ VARCHAR   │
│ youtube_index │ title        │ VARCHAR   │
│ youtube_index │ published_at │ TIMESTAMP │
│ youtube_index │ channel      │ VARCHAR   │
│ youtube_index │ duration     │ BIGINT    │
│ youtube_index │ tags         │ JSON      │
└───────────────┴──────────────┴───────────┘

Exporting data

Export metrics, events, and project metadata to Parquet or CSV for use in other tools:

velocirepo export ./out/

This writes one file per table:

  out/metrics.parquet (257 KB)
  out/github_events.parquet (2.9 MB)
  out/youtube_index.parquet (15 KB)
  out/indicators.parquet (42 KB)
  out/projects.parquet (2 KB)

Use --format csv for CSV output, and --source or --project to filter:

velocirepo export ./out/ --format csv
velocirepo export ./out/ --source github
velocirepo export ./out/ --project quarto

Using the data with other tools

velocirepo provides two ways to access your metrics data from external tools:

1. DuckDB file — a persistent velocirepo/data/velocirepo.duckdb file with views over the raw JSONL data and a projects table. Open it directly from any tool that supports DuckDB.
2. Parquet export — use velocirepo export ./out/ to write Parquet (or CSV) files that any data tool can read.

The DuckDB file is rebuilt automatically after every fetch and project mutation command. You can also rebuild it manually:

velocirepo build-db

Note: The DuckDB file uses relative paths, so external tools must open it from the data directory (or set DuckDB's working directory to it).

Tool compatibility

Tool / Package	DuckDB file	Parquet
DuckDB CLI	Yes	Yes
Python duckdb	Yes	Yes
Polars (Python/Rust)	No	Yes
pandas	No	Yes
Marimo	Yes (via duckdb)	Yes
R duckdb / DBI	Yes	Yes
R arrow	No	Yes
Observable / Evidence	Yes	Yes
Excel / Google Sheets	No	Yes (via CSV)

Examples

DuckDB CLI:

cd velocirepo/data
duckdb velocirepo.duckdb "SELECT project, SUM(value) AS stars FROM metrics WHERE metric = 'daily_stars' GROUP BY project ORDER BY stars DESC"

Python (duckdb):

import duckdb
import os

os.chdir("velocirepo/data")
con = duckdb.connect("velocirepo.duckdb", read_only=True)
df = con.sql("SELECT * FROM metrics WHERE source = 'pypi' ORDER BY date DESC LIMIT 10").df()

Polars (from Parquet export):

import polars as pl

metrics = pl.read_parquet("out/metrics.parquet")
stars = metrics.filter(pl.col("metric") == "daily_stars").group_by("project").agg(pl.col("value").sum())

Indicators

The indicators view computes derived signals from your daily metrics, giving you a sense of how fast a project is growing and in which direction it's heading. Indicators are computed using 28-day trailing windows and are available wherever you query data — via velocirepo query, the persistent .duckdb file, and Parquet exports.

Only metrics with a daily_ prefix are included (these represent per-day deltas like daily_stars, daily_downloads, daily_pageviews). Cumulative total_* metrics are excluded since growth rates on snapshots aren't meaningful.

Schema

Column	Type	Description
project	VARCHAR	Project ID
source	VARCHAR	Source name (github, pypi, etc.)
target	VARCHAR	Specific package, repo, or site being tracked
metric	VARCHAR	Underlying metric (e.g., daily_stars)
indicator	VARCHAR	Indicator name (growth_rate or trend)
date	DATE	Date of computation
value	DOUBLE	Computed value
tags	JSON	Optional metadata from the underlying metric

Growth rate

Measures how much activity increased or decreased compared to the prior period. Computed as:

growth_rate = (sum_last_28d - sum_prior_28d) / sum_prior_28d

A value of 0.15 means 15% more activity in the last 28 days compared to the 28 days before that. Negative values indicate declining activity.

velocirepo query "
  SELECT project, metric, date, ROUND(value, 3) AS growth_rate
  FROM indicators
  WHERE indicator = 'growth_rate'
    AND metric = 'daily_stars'
  ORDER BY date DESC
  LIMIT 5
"

┌──────────────┬─────────────┬────────────┬─────────────┐
│   project    │   metric    │    date    │ growth_rate │
├──────────────┼─────────────┼────────────┼─────────────┤
│ py-shiny     │ daily_stars │ 2026-06-22 │ 0.114       │
│ dplyr        │ daily_stars │ 2026-06-22 │ -0.182      │
│ plotnine     │ daily_stars │ 2026-06-22 │ -0.488      │
│ great-tables │ daily_stars │ 2026-06-22 │ -0.174      │
│ ggsql        │ daily_stars │ 2026-06-22 │ -0.875      │
└──────────────┴─────────────┴────────────┴─────────────┘

Trend

Measures the daily rate of change via linear regression over the trailing 28 days. The value represents units per day — for example, a trend of 3.2 on daily_stars means the project is gaining roughly 3.2 more stars per day than it was at the start of the window.

velocirepo query "
  SELECT project, metric, date, ROUND(value, 2) AS trend_per_day
  FROM indicators
  WHERE indicator = 'trend'
    AND metric = 'daily_downloads'
    AND project = 'plotnine'
  ORDER BY date DESC
  LIMIT 5
"

┌──────────┬─────────────────┬────────────┬───────────────┐
│ project  │     metric      │    date    │ trend_per_day │
├──────────┼─────────────────┼────────────┼───────────────┤
│ plotnine │ daily_downloads │ 2026-06-22 │ -140.39       │
│ plotnine │ daily_downloads │ 2026-06-21 │ -60.35        │
│ plotnine │ daily_downloads │ 2026-06-20 │ 135.05        │
│ plotnine │ daily_downloads │ 2026-06-19 │ 186.6         │
│ plotnine │ daily_downloads │ 2026-06-18 │ 112.38        │
└──────────┴─────────────────┴────────────┴───────────────┘

Querying indicators

You can join indicators with project metadata for richer views:

velocirepo query "
  SELECT p.name, i.metric, i.indicator, i.date, ROUND(i.value, 3) AS value
  FROM indicators i
  JOIN projects p ON i.project = p.id
  WHERE i.date = (SELECT MAX(date) FROM indicators)
  ORDER BY i.indicator, i.value DESC
  LIMIT 5
"

┌────────────┬─────────────────────┬─────────────┬────────────┬───────┐
│    name    │       metric        │  indicator  │    date    │ value │
├────────────┼─────────────────────┼─────────────┼────────────┼───────┤
│ great-docs │ daily_issues_closed │ growth_rate │ 2026-06-22 │ 5.143 │
│ mcp-repl   │ daily_prs_merged    │ growth_rate │ 2026-06-22 │ 3.143 │
│ mcp-repl   │ daily_prs_opened    │ growth_rate │ 2026-06-22 │ 2.815 │
│ great-docs │ daily_issues_opened │ growth_rate │ 2026-06-22 │ 1.647 │
│ Positron   │ daily_pr_comment    │ growth_rate │ 2026-06-22 │ 1.271 │
└────────────┴─────────────────────┴─────────────┴────────────┴───────┘

Custom indicators

You can define your own indicators in velocirepo.toml. Each indicator is a named TOML table with a description and a SQL query:

[indicators.weekly_momentum]
description = "7-day trailing sum"
query = """
  SELECT project, source, target, metric,
    '{{indicator_name}}' AS indicator, date,
    SUM(value) OVER w AS value,
    tags
  FROM metrics WHERE metric LIKE 'daily_%'
  WINDOW w AS (PARTITION BY project, source, target, metric, tags ORDER BY date ROWS 6 PRECEDING)
"""

The {{indicator_name}} placeholder is replaced with the TOML key at runtime (e.g., weekly_momentum).

Behavior: Custom indicators are added alongside the built-in defaults. To disable the defaults, set:

[settings]
include_default_indicators = false

To inspect the built-in defaults:

velocirepo show-indicators --defaults

Without --defaults, show-indicators prints all active indicators (defaults + custom). You can redirect to your config to use as a starting point:

velocirepo show-indicators --defaults >> velocirepo.toml

Generating badges

Generate shields.io-style SVG badges from your metrics data:

velocirepo badge stars -o stars.svg
velocirepo badge forks --project quarto -o forks.svg
velocirepo badge downloads --project ggplot2 -o downloads.svg

For custom metrics, provide a SQL query that returns a single value:

velocirepo badge custom \
  --label "contributors" \
  --query "SELECT COUNT(DISTINCT \"user\") AS value FROM github_events" \
  --color "#ea7233" \
  -o contributors.svg

Styles

Three styles are available via --style:

Style	Description
flat (default)	Subtle gradient, rounded corners
flat-square	No gradient, sharp corners
plastic	Heavy gradient, more rounded

Options

Flag	Default	Description
--project	all	Scope to a specific project
--style	flat	Badge style
--color	preset-dependent	Message background color (hex)
--label-color	#555	Label background color
--label	preset-dependent	Override label text
--height	style-dependent	Badge height in pixels
--radius	style-dependent	Corner radius (0 = square)
-o	stdout	Output file path

Numbers are automatically formatted for readability (e.g., 5274 becomes 5.3k, 1500000 becomes 1.5M).

Views

Views let you create dashboards, reports, and visualizations from your metrics data using your preferred framework. Velocirepo handles scaffolding, data export, and render orchestration — the actual rendering is done by external tools.

Supported frameworks

Framework	Extension	Renderer	Serve mode
Quarto	.qmd	quarto	quarto preview
Jupyter	.ipynb	jupyter	jupyter notebook
Marimo	.py	python	marimo edit
R	.R	Rscript	render + open
ggsql	.sql	ggsql	render + open

Quick start

# Scaffold a view
velocirepo add-view stars --framework sql

# Edit the generated file (views/stars.sql)
# Then render it
velocirepo render-view stars

# Or start a live dev server (Quarto, Marimo, Jupyter)
velocirepo serve-view overview

Data sources

Views can read data in two ways, controlled by the --source flag (or [views].source in config):

• parquet (default): Reads from exported Parquet files in views/_data/. Faster queries, includes computed views like metrics with aggregated GitHub events. The render-view / render-views commands auto-export before rendering.
• jsonl: Reads raw JSONL files directly via DuckDB's read_json_auto(). Simpler but lacks computed views.

Configuration

Add a [views] section to velocirepo.toml for global settings. Per-view overrides go in [[views.items]]:

[views]
dir = "views"          # default: velocirepo/views
source = "parquet"     # default data source for new views

[[views.items]]
path = "overview.py"
venv = ".venv"         # use a specific Python venv

[[views.items]]
path = "raw-debug.qmd"
source = "jsonl"       # this view reads raw JSONL

Views not listed in [[views.items]] still work with defaults — config entries are only needed for overrides (venv, output path, source).

Rendering

velocirepo render-view stars         # render one view
velocirepo render-views              # render all views
velocirepo render-views weekly/      # render views matching a prefix
velocirepo render-views --no-export  # skip Parquet export step

Output goes to views/_output/ by default, mirroring the source tree structure.

CI usage

- name: Render views
  run: velocirepo render-views

- name: Commit rendered output
  run: |
    git add views/_output/
    git diff --staged --quiet || git commit -m "Update rendered views"

MCP server

velocirepo includes a built-in Model Context Protocol (MCP) server, allowing AI assistants like Claude to query your metrics, trigger fetches, and manage projects conversationally.

Starting the server

velocirepo mcp                          # stdio (for Claude Desktop / Claude Code)
velocirepo mcp --http 127.0.0.1:8080    # Streamable HTTP
velocirepo mcp --read-only              # Disable fetch/write tools

Claude Desktop configuration

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "velocirepo": {
      "command": "velocirepo",
      "args": ["mcp", "--config", "/path/to/velocirepo.toml"]
    }
  }
}

Claude Code configuration

Add to your project's .mcp.json:

{
  "mcpServers": {
    "velocirepo": {
      "command": "velocirepo",
      "args": ["mcp"]
    }
  }
}

Available tools

Tool	Description
query	Run SQL against metrics, github_events, youtube_index, projects views
schema	Show all table columns and types
fetch	Fetch from all configured sources
fetch_github	Fetch GitHub events
fetch_traffic	Fetch GitHub traffic
fetch_pypi	Fetch PyPI downloads
fetch_cran	Fetch CRAN downloads
fetch_homebrew	Fetch Homebrew installs
fetch_plausible	Fetch Plausible analytics
fetch_openvsx	Fetch Open VSX metrics
fetch_youtube	Fetch YouTube metrics
list_projects	List configured projects
show_project	Show project details and fetch stats
add_project	Add a project to the config
update_project	Update project configuration
remove_project	Remove a project
rename_project	Rename a project ID
import_projects	Bulk-import from GitHub org/user
validate_projects	Check that source URLs are reachable
list_views	List all views with framework and source
show_view	Show view details
add_view	Scaffold a new view
remove_view	Remove a view
render_view	Render a single view
render_views	Render all or filtered views
badge	Generate an SVG badge
export	Export data to Parquet or CSV
migrate	Migrate data to latest schema
version	Show version info

Example prompts

Once connected, you can ask things like:

• "Which project got the most stars this month?"
• "Show me PyPI download trends for plotnine over the last 6 months"
• "Fetch the latest metrics for all projects"
• "Add a new project called 'my-lib' tracking pypi package my-lib and github owner/my-lib"
• "Generate a stars badge for quarto"

License

MIT