Migration: from hannsxpeter/ready-suite to arc-ready

July 13, 2026 · View on GitHub

If you currently use the eleven-skill hannsxpeter/ready-suite (kickoff-ready, prd-ready, architecture-ready, roadmap-ready, stack-ready, repo-ready, production-ready, deploy-ready, observe-ready, launch-ready, harden-ready) and want to switch to arc-ready, this guide covers the install change, trigger surface, where each former skill's content now lives, and the artifact contract.

The TL;DR: same content, same discipline, same artifacts, one install. The eleven-skill suite remains available; arc-ready is the recommended starting point for new projects.

What changes in arc-ready 1.1

The 1.1 source layout and routing are additive:

SKILL.md is now a concise activation and routing surface. Detailed tier procedures, schemas, examples, and grep tests live in focused references and load only when needed.
Project routing selects a primary form before domain guidance: web application, API or service, CLI or SDK, mobile or desktop, data or ML, or infrastructure or IaC.
Domain guidance composes project form, product archetype, industry overlay, and regulatory overlay. The split domain catalog preserves the inherited guidance while reducing per-task context load.
Public launch preparation can overlap hardening, but public activation requires a fresh .launch-ready/PREPUBLICATION.md tied to the latest hardening revision.
Repository releases now include deterministic evaluations and official Agent Skills validation.

No existing project artifact must move. The frontmatter metadata change and progressive-disclosure refactor affect skill maintainers, not consumer project state.

Install change

Before:

# Eleven separate skills, each in its own repo.
git clone https://github.com/hannsxpeter/kickoff-ready ~/.claude/skills/kickoff-ready
git clone https://github.com/hannsxpeter/prd-ready ~/.claude/skills/prd-ready
git clone https://github.com/hannsxpeter/architecture-ready ~/.claude/skills/architecture-ready
git clone https://github.com/hannsxpeter/roadmap-ready ~/.claude/skills/roadmap-ready
git clone https://github.com/hannsxpeter/stack-ready ~/.claude/skills/stack-ready
git clone https://github.com/hannsxpeter/repo-ready ~/.claude/skills/repo-ready
git clone https://github.com/hannsxpeter/production-ready ~/.claude/skills/production-ready
git clone https://github.com/hannsxpeter/deploy-ready ~/.claude/skills/deploy-ready
git clone https://github.com/hannsxpeter/observe-ready ~/.claude/skills/observe-ready
git clone https://github.com/hannsxpeter/launch-ready ~/.claude/skills/launch-ready
git clone https://github.com/hannsxpeter/harden-ready ~/.claude/skills/harden-ready

After:

# One skill.
git clone https://github.com/hannsxpeter/arc-ready ~/.claude/skills/arc-ready

For other harnesses (Codex, Cursor, Windsurf, Antigravity, Pi, OpenClaw), install per the harness's Agent Skills standard install path.

Trigger surface

The eleven-skill suite had per-skill triggers ("write a PRD" routed to prd-ready; "deploy this" routed to deploy-ready). arc-ready inherits the union of every named trigger and routes internally to the right tier sub-step.

Trigger phrase	Former skill	Now in arc-ready
kickoff, walk me through idea to launch, orchestrate the whole arc	kickoff-ready	Tier 0 + full Mode A dispatch
write a PRD, product spec, requirements doc	prd-ready	Tier 1.1
design the architecture, system diagram, ADR	architecture-ready	Tier 1.2
build a roadmap, milestone plan, Now-Next-Later	roadmap-ready	Tier 1.3
what stack should I use, pick a database	stack-ready	Tier 1.4
set up a repo, add CI, GitHub Actions	repo-ready	Tier 2.1
dashboard, admin panel, internal tool	production-ready	Tier 2.2
deploy this, CI/CD pipeline, expand-contract	deploy-ready	Tier 3.1
add monitoring, define an SLO, write a runbook	observe-ready	Tier 3.2
launch my product, build a landing page, Product Hunt	launch-ready	Tier 3.3
adversarial review, pen-test prep, OWASP walkthrough	harden-ready	Tier 3.4

The activation summary is in the SKILL.md description; expanded examples are in README.md.

Where former skill content lives now

Each former skill's content is preserved in arc-ready under the corresponding tier folder.

kickoff-ready

Former path	New path
`kickoff-ready/SKILL.md`	Routed by `SKILL.md`; detailed Tier 0 procedures live in focused `references/orchestration/` files
`kickoff-ready/references/kickoff-antipatterns.md`	`references/orchestration/kickoff-antipatterns.md`
`kickoff-ready/references/sequencing-rules.md`	`references/orchestration/sequencing-rules.md`
`kickoff-ready/references/handoff-protocols.md`	`references/orchestration/handoff-protocols.md`
`kickoff-ready/references/progress-tracking.md`	`references/orchestration/progress-tracking.md`
`kickoff-ready/references/scope-fence.md`	`references/orchestration/scope-fence.md`
`kickoff-ready/references/agents-md-template.md`	`references/orchestration/agents-md-template.md`
`kickoff-ready/references/RESEARCH-2026-04.md`	`references/shared/RESEARCH-2026-04.md` (consolidated)

prd-ready, architecture-ready, roadmap-ready, stack-ready

Former path pattern	New path pattern
`<planning-skill>/SKILL.md`	Routed by `SKILL.md`; detailed procedures live in `references/planning/planning-workflow.md`
`<planning-skill>/references/*.md`	`references/planning/*.md` (filenames preserved)
`<planning-skill>/references/RESEARCH-2026-04.md`	`references/shared/RESEARCH-2026-04.md` (consolidated)
`<planning-skill>/references/EXAMPLE-*.md`	`references/planning/EXAMPLE-*.md`

repo-ready, production-ready

Former path pattern	New path pattern
`<building-skill>/SKILL.md`	Routed by `SKILL.md`; detailed procedures live in `references/building/building-workflow.md`
`<building-skill>/references/*.md`	`references/building/*.md` (filenames preserved)
`production-ready/ORCHESTRATORS.md`	`references/shared/ORCHESTRATORS.md`

deploy-ready, observe-ready, launch-ready, harden-ready

Former path pattern	New path pattern
`<shipping-skill>/SKILL.md`	Routed by `SKILL.md`; detailed procedures live in `references/shipping/shipping-workflow.md`
`<shipping-skill>/references/*.md`	`references/shipping/*.md` (filenames preserved)
`<shipping-skill>/references/RESEARCH-2026-04.md`	`references/shared/RESEARCH-2026-04.md` (consolidated)

Hub (hannsxpeter/ready-suite)

Former path	New equivalent
`ready-suite/SUITE.md`	Removed. arc-ready is one repo; the byte-identical SUITE.md ritual does not apply.
`ready-suite/ORCHESTRATORS.md`	`references/shared/ORCHESTRATORS.md`
`ready-suite/MAINTAINING.md`	`MAINTAINING.md` (rewritten for single-repo rituals; the v2.5.12 precedent recovery story and multi-repo coordinated-patch matrix are removed).
`ready-suite/scripts/lint.sh`	`scripts/lint.sh` (rewritten for single-repo checks; suite-md-sync removed; trigger-overlap removed).
`ready-suite/references/TRIGGER-DISAMBIGUATION.md`	Folded into `SKILL.md` mode-detection and tier-dispatch sections; retained as `references/orchestration/trigger-disambiguation.md`.
`ready-suite/install.sh`, `uninstall.sh`	Not needed. Standard Agent Skills install applies.

Artifact contract: unchanged

The most important property of the migration is that artifact paths are identical. Downstream consumers (orchestrators like GSD, BMAD, Spec Kit, Superpowers; the dogfood at hannsxpeter/ready-suite-example) consume artifacts at:

.prd-ready/PRD.md
.architecture-ready/ARCH.md
.roadmap-ready/ROADMAP.md
.stack-ready/STACK.md
.repo-ready/SCAFFOLD.md (or .repo-ready/AUDIT-REPORT.md for Mode B audits)
.production-ready/STATE.md
.deploy-ready/DEPLOY.md
.observe-ready/OBSERVE.md
.launch-ready/STATE.md
.launch-ready/PREPUBLICATION.md (required companion only when public activation is in scope)
.harden-ready/FINDINGS.md

These paths are stable across the eleven-skill suite and arc-ready. arc-ready also writes its own .arc-ready/PROGRESS.md for the cross-tier ledger.

arc-ready file-system projects also get Pillars-compatible agent memory:

AGENTS.md with the Pillars loading protocol and an arc-ready artifact map.
agents/context.md and agents/repo.md as mandatory floor pillars.
Additional source-backed agents/*.md files when the relevant arc artifacts exist.

The arc artifacts remain authoritative. Pillars is the project-memory layer for future agent work.

Pillars adoption during migration

For projects created under the eleven-skill suite, the migration path is additive:

Keep every existing .<tier>-ready/ artifact in place.
Add .arc-ready/PROGRESS.md to record imported tier state.
If AGENTS.md is absent, emit the Pillars-compatible loader and the agents/context.md and agents/repo.md floor pillars.
If AGENTS.md already exists and is Pillars-compatible, preserve it and add missing arc-ready artifact-map context only where it fits cleanly.
If AGENTS.md exists and conflicts with Pillars, leave it untouched and record pillars: adoption-blocked-existing-agents in .arc-ready/PROGRESS.md.

Do not rewrite old arc artifacts into Pillars. Pillars points future agents toward the right decisions; the artifact files remain the decision record.

Discipline: unchanged

Every named failure mode, every grep test, every workflow guard from the source eleven skills is preserved in arc-ready. The full per-tier catalogs (with citations, severity, remediation) are at:

references/orchestration/kickoff-antipatterns.md
references/planning/prd-antipatterns.md
references/planning/architecture-antipatterns.md
references/planning/roadmap-antipatterns.md
references/planning/stack-antipatterns.md
references/building/repo-antipatterns.md
references/building/production-antipatterns.md
references/shipping/deploy-antipatterns.md
references/shipping/observe-antipatterns.md
references/shipping/launch-antipatterns.md
references/shipping/harden-antipatterns.md

The load-bearing pattern catalog and grep tests remain available through references/orchestration/failure-mode-catalog.md and references/orchestration/verification-grep-tests.md, with per-tier antipattern references preserving each inherited pattern.

Running both

The eleven-skill suite and arc-ready can coexist on the same machine. The trigger surface overlaps; the harness will route to whichever skill matches first. To prefer arc-ready on a project, ensure arc-ready is installed and the per-skill suite repos are not installed for that project.

There is no harm in keeping both around during a transition period. The artifact contracts are the same, so a project kicked off under the eleven-skill suite is fully consumable by arc-ready (Mode A or B with imports), and vice versa.

When to switch

Switch when:

You are starting a new project (the install footprint difference is most visible here).
You are onboarding a new team member (one skill to install instead of eleven).
You are integrating with an orchestrator (downstream consumer needs only one trigger to route).

Stay on the eleven-skill suite when:

You have customized one or more individual skills' references and the per-skill repo is your version-control surface.
You are running a CI lint that depends on the byte-identical SUITE.md ritual across suite repos.
You prefer the per-skill changelog granularity (eleven CHANGELOG.md files, each tracking one skill's evolution).

Both are valid. arc-ready is the recommended path; the eleven-skill suite is the supported alternative.

Open questions

Trigger collision: if both arc-ready and the eleven-skill suite are installed, the harness may route ambiguously. Recommended posture: install one or the other, not both, on a given machine.
Dogfood maintenance: hannsxpeter/ready-suite-example continues to dogfood the eleven-skill suite. arc-ready's dogfood acceptance test is that the same artifacts (at the same paths) are reproducible from arc-ready. If they diverge, the dogfood is the authority.
Pillars adoption for imported projects: a project started under the eleven-skill suite may not have agents/*.md. When arc-ready imports it, Tier 2.1 should add the Pillars memory layer unless an existing AGENTS.md blocks adoption.