DigestAI news desk

Cut through the AI noise.

Agents & Tools3 min read

Agent-net launches Webagent, open‑source Go harness for turning websites into AI agents

Agent-net, the team behind an agent‑to‑agent marketplace, has open‑sourced Webagent, a Go‑based framework that lets a business expose its website as a public‑facing AI agent. Users supply a single declarative JSON spec, choose providers for nine defined slots, and run a single command to spin up the agent.

1 source

Key points

  • Webagent lets businesses turn a website into a public‑facing AI agent using a single JSON spec and nine pluggable slots.
  • The harness runs out‑of‑the‑box with OpenRouter or gateway LLM brains and supports Slack, WhatsApp, and HTTP channels.
  • Safety is enforced in code via action.Guard, ensuring every tool call passes a guardrail before execution.

The current release (v0, Apache 2.0) includes ready‑to‑use integrations with OpenRouter or gateway LLM “brains” and channel adapters for Slack, WhatsApp, and generic HTTP. Each tool the agent can invoke is wrapped by an action.Guard, a code‑level safety layer that enforces guardrails before execution. While core functionality works, browser actions, OAuth‑gated MCP, OpenTelemetry export, and billing are still on the roadmap.

Webagent aims to prove that robust architecture—not just model size—drives agent success, echoing research that shows an 85 % versus 50 % task‑completion gap when the same models are placed in a well‑engineered agent stack. The project invites businesses to extend the platform by registering custom providers, while partner companies can ship adapters that meet the strict conformance suite.

Full story from MarkTechPost · by Michal SutterOpen source ↗

Agent-net Open Sources Webagent: A Go Harness That Turns Any Website into a Guarded AI Agent

MarkTechPost · 15 September 2026

Loading the full article…

This text was published by MarkTechPost and written by Michal Sutter. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Agents & Tools

All →

Related stories