Skip to main content

Fagun — one MCP server that lets any AI tool (Claude, Cursor, Codex, Antigravity, Windsurf) drive a browser, hunt bugs, and run a full QA sweep.

Project description

🦊 Fagun

Give any AI a browser and let it hunt real bugs for you.

Fagun is a single tool that plugs into Claude, Cursor, Codex, Antigravity, Windsurf, Cline, or VS Code. Once it's set up, you just type fagun (or /fagun) and your AI can open a real browser, click around, and run a full quality check on any website — finding broken links, console errors, failed requests, form problems, accessibility issues, slow pages, and security misconfigurations.

You set it up once. It works in every AI tool. Chrome installs itself.


🚀 Install (any OS — no Python or pip needed)

Step 1 — install uv (it brings its own Python, so nothing else is required):

macOS / Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows (PowerShell):

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

⚠️ Restart your terminal after this so uv is on your PATH. (macOS/Linux: or run source $HOME/.local/bin/env in the current shell.)

Step 2 — set up Fagun:

uvx fagun setup      # installs the Chrome engine automatically
uvx fagun install    # shows the config to paste into your AI tool

That's it. Restart your AI tool, type fagun, and go.

Already have pip/Python? pip install uv also works — but the installer above needs no Python at all, which is why we recommend it.

💡 Don't want to think about config? Just tell your AI: "Install and set up fagun for me — follow https://github.com/mejbaurbahar/fagun/blob/main/install.md" and it does everything above for you.


🔌 Connect it to your AI tool

Every tool uses the same setting: run uvx fagun. Pick yours:

Tool How
Claude Code claude mcp add fagun -- uvx fagun
Claude Desktop add the JSON below to claude_desktop_config.json
Cursor uvx fagun install cursor (writes ~/.cursor/mcp.json)
VS Code (Copilot) uvx fagun install vscode (writes .vscode/mcp.json)
Windsurf / Cline / Antigravity paste the JSON below into their MCP settings
Codex CLI add the TOML below to ~/.codex/config.toml
// Claude Desktop / Cursor / Windsurf / Cline / Antigravity
{ "mcpServers": { "fagun": { "command": "uvx", "args": ["fagun"] } } }
# Codex — ~/.codex/config.toml
[mcp_servers.fagun]
command = "uvx"
args = ["fagun"]

Restart the tool after adding it. Then type fagun.


💬 How to use it

Just talk to your AI in plain English:

  • fagun → shows the menu and starts up
  • go to example.com and take a screenshot
  • run QA on https://example.com
  • deep test https://example.com and save the report to ./report.md
  • check for broken links on https://example.com
  • test the forms on the signup page
  • are there any console errors? · any failed network requests?
  • log in with test@x.com / password123, then check the dashboard

🕵️ The /fagun bug hunter

Fagun ships with a skill that turns your AI into a methodical QA tester. It sweeps 10 kinds of problems and only reports bugs it can actually reproduce (no guessing):

# Checks for
1 Functional — broken journeys, buttons/links that lie
2 JavaScript errors — crashes, console errors on load & on click
3 Network / API — 4xx/5xx, failed calls, mixed content
4 Forms — missing validation, insecure submission, no labels
5 Auth / sessions — login errors, leaks, access control
6 Accessibility — missing alt text, labels, keyboard traps
7 Performance — slow loads, heavy resources
8 Visual / responsive — layout breakage, overflow, cut-off text
9 Security — missing CSP/HSTS, exposed versions, secrets in code
10 Edge cases — reloads, back button, huge inputs, offline

Every finding comes with steps to reproduce, what was observed, and the impact.


⚙️ Options (optional)

Set these as environment variables if you need them:

Variable Default What it does
FAGUN_HEADLESS 1 Set to 0 to watch the browser work
FAGUN_BROWSER chromium Use firefox or webkit instead
FAGUN_CDP_URL Attach to your own open Chrome, e.g. http://127.0.0.1:9222

🧰 Everything it can do (MCP tools)

fagun_start · open_browser · navigate · click · fill · press_key · screenshot · evaluate_js · get_console · get_network · crawl · run_qa · check_links · test_forms · security_headers · deep_test · full_qa_sweep · write_report · close_browser


🛠️ For developers

git clone https://github.com/mejbaurbahar/fagun && cd fagun
pip install -e .
python -m playwright install chromium
python -m fagun        # runs the MCP server on stdio

Release (maintainer): publishing is automatic via GitHub Actions + PyPI Trusted Publishing. Bump the version in pyproject.toml and src/fagun/__init__.py, then:

git tag v0.3.0 && git push origin v0.3.0

MIT © Mejbaur Bahar Fagun

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fagun-0.3.1.tar.gz (19.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

fagun-0.3.1-py3-none-any.whl (17.1 kB view details)

Uploaded Python 3

File details

Details for the file fagun-0.3.1.tar.gz.

File metadata

  • Download URL: fagun-0.3.1.tar.gz
  • Upload date:
  • Size: 19.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for fagun-0.3.1.tar.gz
Algorithm Hash digest
SHA256 dcb3af1576d7c3c4a1d81510d568b62bc85ba15e8eed9f37a52a3d309785ebf4
MD5 bd8170c66033674cff6627756e76d462
BLAKE2b-256 a007d8d7d9df6d0b0b08e5bacef22bd6ff7a901be5e2b4c4af02b12180fef469

See more details on using hashes here.

Provenance

The following attestation bundles were made for fagun-0.3.1.tar.gz:

Publisher: publish.yml on mejbaurbahar/fagun

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file fagun-0.3.1-py3-none-any.whl.

File metadata

  • Download URL: fagun-0.3.1-py3-none-any.whl
  • Upload date:
  • Size: 17.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for fagun-0.3.1-py3-none-any.whl
Algorithm Hash digest
SHA256 e645820ab9f1ec5343cee5f3a34ccb819dec4cc373ddfcbb54a73f8fbb853b3f
MD5 10bb9ceb918295a19673f30f6fe90ed4
BLAKE2b-256 5fe1a067120eb9005b507e97ba897890666e3b26fd99d0db02333c30b3079058

See more details on using hashes here.

Provenance

The following attestation bundles were made for fagun-0.3.1-py3-none-any.whl:

Publisher: publish.yml on mejbaurbahar/fagun

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page