@everyx/pi-web-tools

Web primitives for pi — search and URL fetch (URL → Markdown). Multi-provider search with keyless-first quota escalation and a real-browser fallback fuse.

Packages

Package details

extension

Install @everyx/pi-web-tools from npm and Pi will load the resources declared by the package manifest.

npm repo report

$ pi install npm:@everyx/pi-web-tools

Package: @everyx/pi-web-tools
Version: 1.2.4
Published: Sep 11, 2026
Downloads: 1,482/mo · 243/wk
Author: every.x
License: MIT
Types: extension
Size: 116.1 KB
Dependencies: 6 dependencies · 3 peers

Pi manifest JSON

{
  "extensions": [
    "./index.ts"
  ]
}

Security note

Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.

README

pi-web-tools

English | 中文

Web primitives for Pi — web_search + web_fetch.

web_search — search the web. Requests route across HTTP search providers in fuse order — TinyFish (free, unlimited) → Exa → Tavily → Firecrawl (Google-backed; keyless first, key only as escalation) — with automatic failover; a real-browser engine (Google, Baidu for Chinese via BrowserSkill) is the last-resort no-key fuse. Five params: query, recency?, allowed_domains?, blocked_domains?, locale?. Results follow the query's language; locale is the explicit market/language boost. No engine param — routing is invisible to the LLM.
web_fetch — fetch a URL: HTML pages convert to Markdown by default (LLM / token friendly), pass raw: true to get the source HTML as-is. Non-HTML (SVG/JSON/text — anything textual) is always returned as-is; oversized non-web content is stashed to /tmp with a pointer instead of an inline preview (the LLM pages through it via read). Images (image/*, except SVG) are returned as multimodal image blocks — rendered inline in the TUI and consumed by the model directly. The response Content-Type travels in the result metadata. Direct fetches wake the system curl CLI (identity = a real curl of this machine — TLS, headers, HTTP/1.1 all native, nothing mimicked), with a pinned-curl-UA fallback on curl-less systems; anti-bot / JS-rendered pages advance to the fuse (tinyfish fetch → real browser).

Two primitives, nothing else — no content storage, no curator, no PDF/video extraction.

Install

pi install npm:@everyx/pi-web-tools

Usage

Search the web for the pi extension API docs

Fetch https://github.com/earendil-works/pi/blob/main/README.md and summarize it

Configuration

Variable	Meaning
`TINYFISH_API_KEY`	Enable the TinyFish channel (free, unlimited volume, 30 req/min) — the primary channel.
`EXA_API_KEY`	Exa keyed mode (full params). Without it Exa still runs keyless via MCP (bare queries, 3 qps / 150 calls per day).
`TAVILY_API_KEY`	Enable the Tavily channel (1,000 credits/month free).
`FIRECRAWL_API_KEY`	Firecrawl keyed escalation. Without it the channel runs keyless (free, per-IP daily caps). Never touched while keyless quota lasts — the pool is shared with pi-read-doc's OCR.

Zero keys still works: keyless Exa + keyless Firecrawl, with the real-browser fuse behind them. The real-browser channel uses BrowserSkill (bsk CLI); install it separately.

Notes

Not compatible with pi-web-access: both register a web_search tool. Uninstall one of them.
Design intent and requirements: SPEC.md.