Built with Hugo and WordPress, hosted on Fastly. Domain registered in 2026.
adrs.systems is built with Hugo, hosted on Fastly.
| Technology | Version | First seen |
|---|---|---|
| CMS & Site Builders2 | ||
| 0.154.5 | 6 Nov 2026 | |
| – | 6 Nov 2026 | |
| WordPress Plugins1 | ||
| – | 6 Nov 2026 | |
| Web Servers, Languages & Backend Frameworks1 | ||
| – | 6 Nov 2026 | |
| CDN & Hosting3 | ||
| – | 6 Nov 2026 | |
Registration from the registry's RDAP record, the certificate the site serves, and its rank in public traffic lists.
The crew hasn’t read this site’s colors yet. We read a palette when we capture a site’s screenshot, busiest sites first. Meanwhile, browse palettes from other sites.
Counted from every stack change we've recorded: of the sites we watched running one of this site's tools, how many went on to add another. A pattern across sites, not a forecast for this one.
mx2.efwd.spaceship.netmx1.efwd.spaceship.netlaunch1.spaceship.netlaunch2.spaceship.netv=spf1 include:spf.efwd.spaceship.net ~all# ADRs for AI > ADRs for AI (adrs.systems) is a method and toolkit for human–AI collaboration in which the agent raises every question it cannot settle in a fixed form — Options, Recommendation, Rationale, Confidence, Falsifier (ORRCF) — and the human's ruling lands as a commit, so every decision is reconstructible from git alone. Architectural Deliberation Records (ADRs), kept in docs/adr/ beside the code they govern, are the record; a template, a Claude Code plugin, an MCP server, a CLI, a VS Code extension, and an iOS client are the tools. This file is the index for models. Each linked resource is machine-readable or plain text where one exists; the human pages are HTML. ## Start here - [Get started: the ten-minute path](https://adrs.systems/get-started/): copy the template, dump your idea, recruit an agent, answer its first question — the flow the founder actually runs. - [What is this](https://adrs.systems/what-is/): the idea in one page. - [ORRCF](https://adrs.systems/orrcf/): the five words every decision question ships with, and why. - [The Handbook (llms.txt)](https://adrs.systems/handbook/llms.txt): the textbook, each page with a markdown twin. ## The record format, for models - [Schemas bundle (llms.txt)](https://schemas.adrs.systems/llms.txt): the current template release as one fetchable teaching bundle — the ADR and seed molds, the QST/ANS grammar, the navigation codes, the skills. - [The template repository](https://github.com/ADRs4AI/human-ai-collaboration-template-A): "Use this template" to start a project; its README teaches the plugin, the MCP server, and the CLI. - [Template README (raw markdown)](https://raw.githubusercontent.com/ADRs4AI/human-ai-collaboration-template-A/main/README.md) - [Template CLAUDE.md (raw markdown)](https://raw.githubusercontent.com/ADRs4AI/human-ai-collaboration-template-A/main/CLAUDE.md): the directives a model follows inside a project. ## Tools - [The adr plugin for Claude Code](https://github.com/ADRs4AI/plugins): `/plugin marketplace add ADRs4AI/plugins` then `/plugin install adr@adrs4ai`; carries the skills, the commands, and the MCP server config. - [@adrs4ai/mcp on npm (registry metadata, JSON)](https://registry.npmjs.org/@adrs4ai/mcp): the MCP server — typed verbs over the record (lookup, list_open_questions, answer_question, apply_batch, …); invoke as `npx -y @adrs4ai/mcp <workspace-root>`. - [@adrs4ai/docket on npm (registry metadata, JSON)](https://registry.npmjs.org/@adrs4ai/docket): the same verbs as a CLI — `npx @adrs4ai/docket overview .` - [ADRs for AI, the VS Code extension](https://marketplace.visualstudio.com/items?itemName=jlumbroso.adrs4ai): tree of open questions, diagnostics, deep links, the bundled MCP server. - [no-adrs-mirror](https://github.com/ADRs4AI/no-adrs-mirror): a GitHub Action that publishes a repository's code without its deliberation record, with an honest notice. - [The ADRs for AI iOS client](https://apps.apple.com/app/id6805589572): review and answer open questions from your phone; an Accept lands as a commit. ## Examples in the open - [frontier-model-observatory](https://github.com/jlumbroso/frontier-model-observatory): a repository where a model made 64 of the first 73 commits under eight ORRCF questions ruled from a phone. - [free-disk-space, ADR-0007](https://github.com/jlumbroso/free-disk-space/blob/main/docs/adr/0007-swap-storage-defaults-to-false-the-policy-acquires-a-boundary.md): decisions reconstructed retroactively on a GitHub Action in daily CI use. - [awesome-adrs4ai](https://github.com/ADRs4AI/awesome-adrs4ai): projects practicing the method, with deliberation-status marks. - [The launch, as a ruled decision](https://github.com/ADRs4AI/community/blob/main/docs/adr/0003-the-launch-announcement-is-itself-a-ruled-decision.md): the announcement's own record, ruling included. ## Papers - [Cybernetic and Epistemic: A Missing Vocabulary for Trustworthy Agentic Delegation](https://arxiv.org/abs/2610.00961): accepted at the AAAI 2026 Fall Symposium on Trustworthy Agentic Systems (TAS 2026); arXiv:2610.00961. - [Deliberative Programming: A Versioned Deliberation Protocol for Human–AI Collectives](https://doi.org/10.5281/zenodo.23300579): DOI 10.5281/zenodo.23300579. - [The methodology template, citable](https://doi.org/10.5281/zenodo.23303469): concept DOI 10.5281/zenodo.23303469 — always resolves to the latest archived release. ## Optional - [Blog](https://adrs.systems/blog/) - [Community: questions and Discussions](https://github.com/ADRs4AI/community) - [Support](https://adrs.systems/support/) · [Privacy](https://adrs.systems/privacy/) - [The GitHub organization](https://github.com/ADRs4AI)
| – |
| 6 Nov 2026 |
| – | 6 Nov 2026 |