Agentability

Agentability is an open weekly experiment where a real AI agent attempts ten everyday web errands.

Agentability is a public project that tests whether AI agents can actually use the web. Every week, an AI agent attempts ten real errands using plain web requests without logins or human help. The project publishes every transcript verbatim, detailing wins and failures. Alongside the show, it audits 113 well-known sites for AI agent readiness, checking conventions like llms.txt, crawler policies, and structured data. Each site receives a public report with a score and fixes for failed checks.

Agentability at a glance

Best for: Software engineer, Product manager
Runs on: Web
Pricing: Free
Released: October 2026
Category: AI agent tool
Website: agentability.org

Key features

Tests AI agents on ten web errands weekly
Publishes verbatim transcripts of agent attempts
Audits 113 well-known sites for agent readiness
Provides public reports with scores and exact fixes
Checks conventions like llms.txt and crawler policies

Use it for

Watch AI agents try and fail at web tasks
Look up a site's agent readiness score
Find fixes to make your own site agent-friendly
Access raw audit data and weekly transcripts

Alternatives to Agentability

BrowserOS — Runs entirely locally and lets AI agents use real user logins to complete tasks.
Not Human Search — Score any site across 7 agentic signals and search 8,000+ ready sites.
Adyen — Share AI agent resources across coding tools and repositories easily.
AgenticSeek — It runs entirely on a local machine without cloud dependency or API costs.
Search2o — Builds model-neutral agents with strict spend limits and role-based access.
Rakazo — Complete ownership of data, models, and execution environment.
Alook — Connects existing local coding agents to team communication channels.
Rinkata — It keeps humans and AI agents aligned on intent as work evolves.
AgentGPT — It allows users to build and deploy autonomous agents in the browser.
Agent Accountability — It applies real consequences and penalties to AI agents for mistakes.
FastGPT — It uses hybrid retrieval and auto-cleansed data to prevent hallucinations.
Composio — Equips AI agents with tools, managed user auth, and sandboxes across 1,500+ apps

All Agentability alternatives