pi-web-minimal

Web research for Pi agents without trashing the context window.

LLMs research badly: a single fetch or search dump can blow 50k tokens of HTML, ads, and nav chrome into context, evicting the actual work. This package wraps Exa + Context7 behind tools that store raw results out-of-band and return a short, source-cited brief. The agent gets evidence; you keep your context budget.

Suckless by design. No browser session, no curator UI, no PDF/video pipeline, no provider zoo.

How

Two-stage pipeline per call:

1. Retrieve via Exa / Context7 / git clone. Raw evidence is stored out-of-band under a responseId; session persistence is bounded so long runs do not bloat context/history.
2. Distill/extract before returning:
   • Small payloads → deterministic extractive compaction (no model call).
   • Larger payloads → your active Pi model runs as a context firewall over ranked snippets: fixed sections, every finding cites [S#], retrieved text treated as untrusted data. Output is validated.
   • If model distillation is unavailable, the fallback is a bounded extractive report, not a first-N raw dump.

You pay one small model call to avoid pasting 50k tokens of HTML into the main context. Override the distiller with PI_WEB_MINIMAL_DISTILL_MODEL=provider/model-id. Set PI_OFFLINE=1 to skip model distillation and use deterministic extraction.

Install

pi install npm:pi-web-minimal
export EXA_API_KEY=exa-...
export CONTEXT7_API_KEY=ctx7sk-...

Or ~/.pi/web-search.json:

{ "exaApiKey": "...", "context7ApiKey": "...", "distillModel": "provider/model-id" }

Tools

| Tool | For | | --- | --- | | web_search | discover current sources | | code_search | API/code examples | | documentation_search | live library docs (Context7) | | fetch_content | URLs + GitHub repos (shallow-cloned to /tmp/pi-github-repos) | | get_search_content | raw escape hatch by responseId with sourceIndex/urlIndex, offset, section, or textSearch selectors when distillation dropped something you needed |

Dev

bun test && bun run typecheck && bun run check

See AGENTS.md for the validation gauntlet, docs/agent-tool-audit.md for design notes.