Metadata-Version: 2.4
Name: pagecheck
Version: 0.1.0
Summary: Pre-launch verdict for a landing page: will it actually capture leads and attribution, or is it silently losing them?
Author-email: Anaum Pandit <anaump7@gmail.com>
License-Expression: MIT
Project-URL: Homepage, https://github.com/panaum/pagecheck
Project-URL: Issues, https://github.com/panaum/pagecheck/issues
Keywords: qa,testing,playwright,landing-page,attribution,lead-capture
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Software Development :: Quality Assurance
Classifier: Topic :: Software Development :: Testing
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: playwright>=1.40
Provides-Extra: web
Requires-Dist: fastapi>=0.110; extra == "web"
Requires-Dist: uvicorn>=0.27; extra == "web"
Requires-Dist: pydantic>=2.0; extra == "web"
Dynamic: license-file

# pagecheck

A landing page can return HTTP 200 on every request and still be losing every
lead. The form posts to an endpoint that no longer exists. The attribution
parameters are stripped by a redirect before anything records them. The submit
button has no destination at all.

`pagecheck` gives a single page a pre-launch verdict: will it actually capture
leads and attribution, or is it silently losing them?

## Install

```bash
pip install pagecheck
playwright install chromium
```

## Use

```bash
pagecheck https://example.com/landing-page
pagecheck --file urls.txt --json
```

Exit code is 0 when nothing failed, 1 when at least one check failed.

## What it does

Loads the page with test attribution parameters attached, reads the forms back,
and reports on lead capture, attribution survival, tracking presence and
destination integrity.

## What it does not do

It is read-only. It never submits a form, never clicks anything that would
create a record, and never writes to the page it is checking.

A check it cannot prove is reported as a warning, not a failure. For a tool
whose output goes to a client, a false alarm costs more than a soft warning.

## Licence

MIT
