Experiment Simulation Presentation — AI skill for Claude Code
Building and simulating a behavioral experiment with an AI agent: a real Claude Code session, console on the left, commentary on the right.
How to install Experiment Simulation Presentation
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open ericcgladstone-maker/experiment-simulation-presentation and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Experiment Simulation Presentation does
Building and simulating a behavioral experiment with an AI agent: a real Claude Code session, console on the left, commentary on the right. Replies and tool output verbatim; the organization and its brief are fictional.
Alternatives in AI
- CodeIsland — Real-time AI coding agent status panel in your MacBook notch — live status, approvals & replies for 13 AI tool 2.3k ★
- Mirofish CLI — Multi-agent AI prediction engine - digital sandbox for scenario simulation (fork of 666ghj/MiroFish) 307 ★
- Vera Eb Suite — Open-source AI skills and plugins for EB-1 and EB-2 NIW immigration petitions — from evidence organization to 219 ★
README
Building and simulating a behavioral experiment with an AI agent
Eric Gladstone · research walkthrough · independent work · September 2026. The presentation's own header reads "Experimental rehearsal · Experimental design".
A real Claude Code session, presented as it was worked through. The Claude Code console is on the left and my commentary is on the right. Starting from a client brief, the session decides what the first experiment has to answer, reviews outside evidence, specifies the measures and decision thresholds, compares designs, and builds the chosen one as protocol, analysis plan and code. It then simulates the experiment 400 times in each of ten synthetic worlds where the truth is known, stress-tests its assumptions, finds where it could mislead a decision, revises it, retests the revision on the same worlds, and ends with a conditional recommendation.
- Live: https://experimentsimulation.eric-c-gladstone.workers.dev
- Also at: https://graystoneindustries.co/talks/ (embedded)
What is verbatim and what is editorial
The questions are mine. Claude's replies and tool output are verbatim, from one continuous session (Claude Code 2.1.282, Opus 5.5), recorded 27 September 2026. Turns 1–20 are the original build. Turns 21–23 are later questions in the same session, labeled on screen, that show one synthetic experiment end to end, draw two figures from the saved results, and bring the decision table forward; they changed nothing the earlier turns produced. Omitted parts of replies are marked on screen. Timing, grouping into parts, focus, and the commentary on the right are editorial.
The organization (Civic Access Network) and its brief are fictional, written for this demonstration. The session ran in an isolated directory, and account names are removed from the published data.
Use
Open it and step with ← →. Use shift+← → to move between sections and space to play. A− / A+ (or the - and = keys) change the terminal text size. URL options: `
Related Skills
Experimental Rehearsal Presentation
Auditing a field-experiment proposal with an AI agent: a real Claude Code session, console on the left, commen
Kivun Terminal Wsl
Claude Code with Hebrew, Arabic and Persian rendered correctly. Most terminals have no bidirectional text engi
Grokkable Output
Agent skill that makes AI replies parseable in one pass — verdict first, prose over fake structure, claims tha
Fidelis
Zero-LLM agent memory for Claude Code and AI agents: local-first BM25, dense-vector, and reciprocal-rank-fusio
Baileys Skill
High-accuracy Baileys knowledge base for AI coding agents (Claude Code, OpenCode, Codex, Antigravity, and more
Jev Compactor
Context compaction and safety gating for AI agents via TypeSafe Jev: keeps messages verbatim, no summarization
Related Agents
Script Writer
Specialist agent for writing speaker notes (presentation scripts) for Quarto RevealJS decks - paper reviews, c
Slides Maker
Use when experiment findings, a research argument, or project onboarding material needs a presentation artifac
Mathodology Coder
Use for reproducible computation, simulation, optimization, figures, tables, and experiment logs.