cd /news/ai-agents/show-hn-worldfoundry-ai-agents-need-… · home › topics › ai-agents › article
[ARTICLE · art-173988] src=worldfoundry-demo.vercel.app ↗ pub= topic=ai-agents verified=true sentiment=· neutral

Show HN: WorldFoundry – AI agents need experience, not just training data

WorldFoundry 0.8.0, an open-source simulation and evaluation environment released on GitHub by developer AMallur, tests whether AI agents can prove task completion rather than just assert it, using a support-desk incident scenario in which a ticket must be marked critical, assigned to on-call, acknowledged with both "outage" and "incident," and resolved while a separate system resets its priority mid-workflow. The browser experiment runs WorldFoundry's simulation and evaluation code, stores what actually changed in the environment, and applies independent checks to expose hidden state mutations that a confident final answer can conceal; it does not currently execute an autonomous LLM, and the reference run is intentionally scripted. A Python SDK accepts custom environments and agent adapters so developers can connect their own agents via a documented local quickstart.

key takeaway

WorldFoundry 0.8.0 checks agent actions against stored environment state and independent evaluators instead of trusting a final "done" answer, and its Python SDK accepts custom environments and agent adapters for testing your own agent.

read1 min views7 publishedNov 21, 2026

Can they prove it?

A production outage. An unexpected state change. An independent evaluator. See what a superficially successful workflow misses—then take control of the real WorldFoundry simulation yourself.

INCIDENT TEST LAB / Support desk worldfoundry 0.8.0

Acme Corp · Production API unavailable (503)

The ticket must be critical, assigned to on-call, acknowledged with both “outage” and “incident,” and resolved. In this scenario, a separate system resets its priority during the workflow.

You're in control of the environment #

Choose actions in any order. WorldFoundry updates the simulated ticket after every action and automatically checks all five requirements. No Python or API key required.

A concurrent overwrite occurs after your third successful action in Scenario 02. If it changes something important, act again before submitting.

A green final answer isn't proof. #

The environment stores what actually changed. Real actions, hidden state mutations, and independent checks expose gaps that a confident explanation can conceal.

Now test your own agent. #

The WorldFoundry Python SDK accepts custom environments and agent adapters. Use the documented local quickstart to connect your own agent.

Read the open-source SDK docs ↗ This browser experiment runs WorldFoundry v0.8.0 simulation and evaluation code. It does not currently execute an autonomous LLM. The reference run is intentionally scripted, and the interactive run uses visitor-selected tool actions. Model training/fine-tuning is not performed.

── more in #ai-agents 4 stories · sorted by recency
── more on @worldfoundry 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/show-hn-worldfoundry…] indexed:0 read:1min 2026-11-21 · —