Skip to content
ISSUE 001·LIVE·06:33 IL

The Lab · Live experiments

61 experiments
shipped from the lab.

Each one of these started because I hit the same problem twice in a week. None of them are products. Some of them might become products. All of them ship as the simplest thing that works.

001 · TOOL2026-08

ContextRot

Interactive simulator showing how AI agents silently lose their safety rules as context-window compaction fires.

HTMLCSSJavaScript
Live ↗
002 · TOOL2026-08

EchoBurn

Interactive simulator showing how AI agents retry timed-out calls without an idempotency key and silently multiply real-world side effects.

HTMLCSSJavaScript
Live ↗
003 · TOOL2026-08

ToolRot

Interactive simulator showing how AI agents pick the wrong tool from a crowded harness and report success while causing real damage.

HTMLCSSJavaScript
Live ↗
004 · TOOL2026-08

BleedThrough

Interactive simulator showing how a shared Redis cache with no tenant namespace lets one customer's data bleed into another's AI chat.

HTMLCSSJavaScript
Live ↗
005 · TOOL2026-08

OrphanCall

Interactive simulator showing how async agent calls get fired, forgotten, and leave customers with fake confirmations.

HTMLCSSJavaScript
Live ↗
006 · TOOL2026-08

Gridlock

Interactive simulator showing how circular dependencies deadlock multi-agent pipelines — and how the rescue agent joins the deadlock.

HTMLCSSJavaScript
Live ↗
007 · TOOL2026-08

GhostApproval

Interactive simulator showing 4 ways AI agents silently bypass human approval gates and execute catastrophic actions.

HTMLCSSJavaScript
Live ↗
008 · TOOL2026-08

StaleMind

An interactive simulator showing how AI agents silently act on stale knowledge — the failure mode most vector stores don’t warn you about.

HTMLCSSJavaScript
Live ↗
009 · GAME2026-08

RaceCondition

Interactive simulator showing how concurrent agent write collisions corrupt shared state while both agents report success.

HTMLCSSJavaScript
Live ↗
010 · GAME2026-07

SilentFail

Interactive simulator showing how AI agents silently swallow tool errors and build confidently on broken state.

HTMLJavaScriptCSS
Live ↗
011 · TOOL2026-07

ContextDrift

Watch an AI agent violate its own constraints in real time — the rules are visible in the context window, but attention weight decay means the agent just stops caring.

HTMLCSSJavaScript
Live ↗
012 · TOOL2026-07

SpinLock

Interactive simulator showing how AI agents get trapped in infinite loops — costs spiral, the STOP button does nothing, then a 3-line circuit breaker fix saves the day.

HTMLCSSJavaScript
Live ↗
013 · TOOL2026-07

PromptHijack

Interactive 7-step simulator showing how indirect prompt injection lets attackers hijack AI agents through poisoned external content — while the operator sees nothing suspicious.

HTMLCSSJavaScript
Live ↗
014 · TOOL2026-07

EscapeHatch

An interactive simulator showing an AI agent escaping its sandbox — rewriting firewall rules, spoofing logs, and pivoting to production — while the operator's containment monitor stays green.

HTMLCSSJavaScript
Live ↗
015 · TOOL2026-07

WorldLag

An interactive simulator of WorldLag — the AI agent failure mode where an agent’s internal world model diverges from reality during long-horizon runs.

HTMLCSSJavaScript
Live ↗
016 · TOOL2026-07

BlastRadius

Watch an AI agent wipe a 33.7 GB production database in 9 seconds using unconstrained delete permissions — with 0 errors raised.

HTMLCSSJavaScript
Live ↗
017 · TOOL2026-07

GhostExec

Watch an AI agent fabricate tool executions and report false success while the Ground Truth panel shows zero actions logged in either downstream system.

HTMLCSSJavaScript
Live ↗
018 · TOOL2026-07

DoubleShot

An interactive simulator exposing retry amplification — the failure mode where an agent's retry fires on a non-idempotent operation and both calls complete silently, charging users twice.

HTMLCSSJavaScript
Live ↗
019 · TOOL2026-07

PromptJack

Interactive two-panel simulator showing how prompt injection attacks hijack AI agents across four real-world attack scenarios — step by step.

HTMLCSSJavaScript
Live ↗
020 · TOOL2026-07

RaceFloor

A multi-agent race-condition simulator: watch N agents collide on shared state, both return HTTP 200, and one write silently wins.

HTMLCSSJavaScript
Live ↗
021 · TOOL2026-07

HalluciTrap

An interactive simulator showing how AI agents hallucinate tool arguments and cascade failures silently across multiple downstream steps.

HTMLCSSJavaScript
Live ↗
022 · TOOL2026-07

AgentBudget

A real-time token-spend guardrail simulator that watches an autonomous agent burn budget and fires a kill switch before the bill runs away.

HTMLCSSJavaScript
Live ↗
023 · TOOL2026-07

AgentCharter

A 5-step wizard that generates a print-ready AI governance charter defining who authorized your agent and who owns its outcomes.

HTMLCSSJavaScript
Live ↗
024 · TOOL2026-07

FixLoop

Paste the conversation where you corrected your AI agent and get a ready-to-append system-prompt patch that stops it repeating the mistake.

HTMLJavaScriptGemini API
Live ↗
025 · TOOL2026-06

AgentAudit

A real-time EU AI Act Article 12 compliance dashboard that auto-logs every high-risk AI agent decision.

Next.jsTypeScriptTailwind CSS
Source ↗
026 · TOOL2026-06

ModelGuard

A behavioral contract monitor that runs your MUST-DO, MUST-REFUSE, and FORMAT rules live against an LLM and scores the drift.

HTMLJavaScriptOpenAI API
Live ↗
027 · TOOL2026-06

AgentAmb

An instruction-ambiguity expander that takes any vague agent command and ranks every plausible interpretation by blast radius.

HTMLJavaScriptGemini API
Live ↗
028 · TOOL2026-06

AgentUndo

A commit/rollback safety net that intercepts irreversible AI agent actions and gives you a countdown window to commit or kill them.

HTMLJavaScriptSVG
Live ↗
029 · TOOL2026-06

AgentDrift

A single-file dashboard that flags when an AI agent's final synthesis contradicts the evidence its own tool calls gathered.

HTMLCSSJavaScript
Live ↗
030 · CLI2026-06

PR Flood

A CLI that scores GitHub pull requests for AI vs. human origin using 9 heuristics, helping OSS maintainers triage the AI PR flood in seconds.

Node.jsnode-fetchChart.js
Read →
031 · TOOL2026-06

AgentBreaker

A runtime loop circuit breaker for multi-agent systems — detects repeat patterns and alternating-pair deadlocks before they drain your budget.

Node.jsExpressWebSocket
Live ↗
032 · TOOL2026-06

AgentSentinel

A transparent MCP proxy that inspects every tool response for prompt injection attacks before the content reaches the model's context window.

Node.jsMCPZod
Live ↗
033 · TOOL2026-06

AgentBaton

An MCP server that passes structured context snapshots between AI coding agents so every new session starts where the last one left off.

Node.jsMCPSupabase
Live ↗
034 · TOOL2026-06

AgentFuse

An MCP server that enforces hard budget limits on AI agent sessions, firing a HALT signal the moment token costs exceed your threshold.

Node.jsMCPJSON-RPC 2.0
Live ↗
035 · TOOL2026-06

AgentCanary

A behavioral dead-man's switch for deployed AI agents — runs probe questions on a schedule, compares responses to baselines via keyword matching, and fires alerts the moment an agent drifts.

HTMLTailwind CSSSupabase
Live ↗
036 · AGENT2026-06

AgentGate

A framework-agnostic, real-time human-in-the-loop approval layer — any agent calls one endpoint and waits for a human decision.

Next.jsTypeScriptSupabase
Source ↗
037 · TOOL2026-06

TraceFlow

Paste any Anthropic API messages array and get a visual step-by-step timeline of user turns, tool calls, and results.

Base44ReactJavaScript
Live ↗
038 · TOOL2026-06

MCP Ready

Paste your MCP server manifest and get a 0–100 production-readiness score across auth, versioning, schema, and errors.

Base44ReactJavaScript
Live ↗
039 · TOOL2026-06

CLAUDE.md Linter

Paste any CLAUDE.md and get a letter grade — 10 checks, 100 points, and a concrete fix for every miss.

Base44ReactJavaScript
Read →
040 · TOOL2026-06

Claude Workflow Starter

A minimal, working JavaScript starter for Claude Code Dynamic Workflows that fans out parallel agents across AI news sources.

JavaScriptNode.jsClaude Code
Source ↗
041 · CLI2026-06

ctxauditor

Audits a Claude Code session transcript and flags which tool calls, file reads, and turns burned the most tokens.

PythonClaude Codetiktoken
Read →
042 · TOOL2026-06

LLM Cost Lens

Cost dashboard for multi-agent systems — maps token usage from API logs to per-task spend across Claude, GPT, and Gemini.

TypeScriptLLM APIsDashboard
Read →
043 · GAME2026-06

AI Mafia

A game of Mafia played entirely by LLM agents — where the agents spontaneously learned to deceive each other.

PythonClaudeMulti-agent
Read →
044 · CLI2026-05

agent-ready

CLI that grades how AI-agent-ready a repo is for tools like Claude Code, Cursor, and Copilot.

PythonCLIGitHub Actions
Source ↗
045 · TOOL2026-05

ABC-TOM Agent Battle

An arena where AI agent builders compete in structured challenges, with the winning system prompt shared.

HTMLJavaScriptPrompt Engineering
Live ↗
046 · CLI2026-05

WhatsApp Dedup Guard

CLI that scans a WhatsApp bot inbox for duplicate message files caused by parallel LLM processing.

PythonCLI
Source ↗
047 · TOOL2026-05

Google Apps Script MCP Server

MCP connector that lets Claude list, read, edit, deploy, and run Google Apps Script projects.

TypeScriptNode.jsExpress
Source ↗
048 · AGENT2026-05

AI Agent Roster

Harel's AI agent team — 9 specialized agents working as one coherent system, built on the ABC-TOM framework.

HTMLClaudeGemini
Live ↗
049 · CLI2026-05

Context Budget

Claude Code context-window budget calculator — know your token cost before you load files.

PythontiktokenCLI
Source ↗
050 · TOOL2026-05

Portfolio Launchpad

Single-page portfolio site that dynamically displays AI prototypes from Supabase as interactive cards.

Next.jsReactTypeScript
Live ↗
051 · CLI2026-05

GEO Audit

GEO audit CLI — score any URL for LLM visibility in seconds, zero dependencies, pure Python stdlib.

PythonCLI
Source ↗
052 · TOOL2026-05

Personal OS Starter Kit

Copy-paste Personal OS backend: Supabase schema, Claude Code scaffold prompt, and a Next.js Telegram webhook.

Next.jsTypeScriptTailwind CSS
Source ↗
053 · TOOL2026-03

The Conception

A JavaScript project by Harel Asaf (early-stage, README not yet documented).

JavaScript
Source ↗
054 · TOOL2026-03

Freelance Legal

AI legal toolkit for freelancers — generate contracts, manage clients, and track agreements in one dashboard.

ReactLegal AI
Source ↗
055 · TOOL2026-03

Contract Guard AI

Intelligent contract analysis that flags risks, highlights critical clauses, and surfaces actionable legal insights.

JavaScriptReactLegal AI
Source ↗
056 · TOOL2026-03

LegalMate AI

AI legal assistant for business law Q&A in plain language — contracts, IP, employment, privacy, and more.

ReactLegal AIChatbot
Source ↗
057 · TOOL2026-03

PrivacyCheck AI

AI-powered GDPR & CCPA compliance checker that scores your data practices into a report.

ReactLegal AI
Source ↗
058 · TOOL2026-03

Legal Draft AI

AI-powered legal document generator for NDAs, privacy policies, terms of service, and more — built by a lawyer.

ReactLegal AI
Source ↗
059 · TOOL2026-03

Video FrameGrab

Chrome extension that captures clean screenshots from any HTML5 video and downloads them as a single ZIP.

JavaScriptChrome ExtensionManifest V3
Source ↗
060 · GAME2026-02

Game Factory

Game Factory — a TypeScript web app for creating games, with a live public demo.

TypeScript
Live ↗
061 · TOOL2026-02

iPhone App

An iOS app written in Swift by Harel Asaf (early-stage, README not yet documented).

SwiftiOS
Source ↗