@gfgghgftgffg/pi-web-access-lean

Part of the amox-pi-extensions-lean monorepo. A lightweight Pi facade for pi-web-access that keeps the full upstream search/fetch engine and compresses four provider tools into one web_access tool. Runtime updates follow upstream.

Original lean facade by kunkun9527; upstream engine by Nico Bailon. Both are MIT-licensed; the original LICENSE is preserved in this package.

Core features

• Routes upstream search, checking, fetching and stored-content tools through web_access, without duplicate registrations.
• Direct local PDF access accepts Unix/Windows paths and file:///, returning the saved Markdown with its path; result metadata reports the Markdown character count and page count when available.
• Advanced parameters on demand: help returns the installed upstream's full description and schema.

Installation

This is a public npm package. Install without a version pin:

pi install npm:@gfgghgftgffg/pi-web-access-lean
pi update --extensions

For a Git fallback, run from the repository checkout:

node scripts/install-extension.mjs pi-web-access-lean

Update a local Git installation by pulling the checkout and rerunning its installer; pi update does not update local contents. Keep one installation source. To migrate, back up settings, remove only the exact old source shown by pi list, then install the npm source.

Do not load this package alongside another pi-web-access wrapper, to avoid duplicate tool registrations.

Usage

{ "op": "search", "input": "Pi coding agent extensions" }

Pass an explicit op and a string input; serialize batch or advanced parameters as a JSON object inside that string, for example:

{ "op": "fetch", "input": "{\"url\":\"https://example.com/page\",\"redirects\":\"same-origin\"}" }

When parameters are unclear, call help with input set to the operation name for the installed runtime's full description and schema. Translate upstream tool hints in results to the corresponding web_access operation; page bodies are not rewritten.

Configuration and limits

Providers, credentials, search routing and fetch options use upstream's web-search.json, normally under ~/.pi/agent/. See the upstream documentation for configuration paths, environment variables, supported services and billing rules. External services may receive queries and page content and incur charges; review their configuration before use.

Direct local PDF access extracts Markdown only, not upstream's structured extract mode; remote PDFs use upstream handling. web_access stays registered directly: upstream dynamic tool-loading settings do not defer or remove this entrypoint.

Context footprint (character proxy)

Fixed ceil(characters / 4), not provider billing tokens.

Development

See the development and maintenance guide and MAINTENANCE.md for adaptation details.