Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
Haoyu Hu avatar

Auto Psynet

  • Updated June 2, 2026
  • Haoyu-Hu/auto-psynet

auto-psynet is a Claude Code marketplace plugin that automates PsyNet behavioral experiment lifecycles—from preregistered plans to deployable code, LLM or human pilots, analysis, and publication-ready papers—for develope

About

auto-psynet (apsy) is a Claude Code plugin that turns a raw research idea into a preregistered plan, working PsyNet experiment code, pilot runs, analysis, and a paper draft. Install with `/plugin marketplace add Haoyu-Hu/auto-psynet` then `/plugin install apsy@auto-psynet`; slash commands like `/apsy:idea`, `/apsy:build`, `/apsy:pilot`, and `/apsy:paper` walk the pipeline. Seven quality gates (G1–G7) enforce preregistration discipline, ethics checks, and human approval before real deployment. An optional MCP server exposes six deterministic tools (`apsy_status`, `apsy_doctor`, `apsy_route`, `apsy_next`, `apsy_power`, `apsy_data_quality`) for Claude Code, Cursor, or Codex. Verified against PsyNet 13.2 and Dallinger 12.2 with Redis and Postgres for local debug.

  • Seven quality gates G1–G7 gate preregistration, build, pilot, deploy, and analysis transitions
  • Slash commands `/apsy:idea` through `/apsy:paper` cover the full experiment lifecycle
  • Optional MCP server exposes 6 engine tools for status, routing, power, and data quality
  • Supports human, LLM-agent, or mixed participants on PsyNet 13.2 / Dallinger 12.2
  • Ethics policy requires explicit human approval and IRB attestation at deploy gate G4

Auto Psynet by the numbers

  • Data as of Jul 7, 2026 (Skillselion catalog sync)
/plugin marketplace add Haoyu-Hu/auto-psynet

Add your badge

Show developers this marketplace is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Last updatedJune 2, 2026
RepositoryHaoyu-Hu/auto-psynet

How do you automate PsyNet experiment pipelines with Claude?

Run `/apsy:idea` and `/apsy:build` to scaffold a PsyNet experiment with preregistered gates before LLM or human pilot deployment.

Who is it for?

Researchers and engineers building PsyNet/Dallinger behavioral experiments who want gated, reproducible automation inside Claude Code.

Skip if: Teams not using PsyNet, Dallinger, or behavioral-experiment infrastructure who only need generic coding assistance.

What you get

Preregistered plan files, PsyNet experiment.py code, pilot data exports, analysis reports, and OSF-ready paper drafts under `$APSY_PROJECT_DIR`.

  • Preregistered experiment plans
  • PsyNet experiment.py projects
  • Analysis reports and paper drafts

By the numbers

  • Enforces 7 quality gates (G1–G7) across the experiment pipeline
  • Optional MCP server exposes 6 deterministic engine tools
  • Verified on PsyNet 13.2 and Dallinger 12.2

Recommended Marketplaces

How it compares

Choose auto-psynet when the repo already targets PsyNet/Dallinger behavioral experiments and you need gated preregistration-to-paper automation rather than generic agent skills.

FAQ

What does Auto Psynet install in Claude Code?

Auto Psynet installs the `apsy` plugin via `/plugin marketplace add Haoyu-Hu/Auto Psynet` and `/plugin install apsy@Auto Psynet`. Setup commands include `/apsy:setup`, `/apsy:install` for PsyNet and Dallinger, and `/apsy:doctor` to verify the runtime.

What are the Auto Psynet quality gates?

Auto Psynet defines seven gates G1–G7 that block pipeline transitions until preregistration, code quality, pilot integrity, human deploy approval, and analysis discipline pass. Gate G4 always requires explicit human approval and IRB attestation before real human studies.

AI Agentsagentsautomationresearch

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.