create-issflow

Scaffold the iStartSoftFlow AI-coding workflow into any project. Stack-agnostic and tool-agnostic. Non-destructive: it never overwrites your files.

Use

cd my-project
npx create-issflow                  # prompts for your AI tool
npx create-issflow --tool=claude    # or pass it: claude | codex | cursor | gemini | aider | all
# then open your tool — Claude Code fires the SessionStart hook automatically
/overview                           # bootstrap the project

Flags:

• --tool=<name> — pick the target tool (skips the prompt). all writes every adapter.
• --ci — also write .github/workflows/issflow-feature.yml (headless /feature via GitHub Actions: label an issue feature:approved → the lane runs).
• --docker — also write Dockerfile.issflow + scripts/feature-docker.js (headless /feature in an unprivileged container; the container is the sandbox). The image must ship Node ≥ 18 — the kit's hooks are Node scripts; the shipped Dockerfile is Node-based and the wrapper preflights node before every run.
• --dry-run — print what would happen, write nothing.
• --force — overwrite existing kit files (default keeps yours; conflicts are written as <file>.issflow-new for you to merge).
• -h, --help — usage.

What it installs

The portable kit (every tool) in <project>/.claude/:

• agents/ — planner · researcher · implementer · test-author · debugger · e2e-runner · synthesizer
• commands/ — /overview /feature /goal /propose /phase /sprint /ui-audit /qa-audit /security-audit /release /uat /change-request /replan /quick /synthesize /runbook /store-wisdom /log-issue /log-decision /unstuck
• skills/ — caveman · grill-me · karpathy-guidelines · ux-design · security · code-standards
• hooks/ — session-start · context-guard · plan-gate (rule-13 enforcement) · pre-compact · subagent-stop · feature-gate (Stop gate for /feature, artifact-verified)
• istartsoft-flow/METHODOLOGY.md — the full methodology (single source of truth)

Plus a root AGENTS.md (the open standard) and the per-tool adapter:

| Tool | Adds | |------|------| | claude | merges hooks into .claude/settings.json | | codex | nothing extra — AGENTS.md is read natively | | cursor | .cursor/rules/ + .cursor/commands/ + .cursor/hooks.json | | gemini | GEMINI.md pointer | | aider | .aider.conf.yml that loads the methodology | | all | every adapter above |

It also un-ignores the workflow dirs in .gitignore if .claude/* was ignored.

Loop

design-research → grill → plan → implement → test → deploy, one vertical slice per phase. Stack-agnostic: declare your stack in docs/OVERVIEW.md; Phase 0 (infra) is N/A when infra is managed. Planning source of truth stays in your PRD / architecture / stories (e.g. BMAD / iSSM).

Maintainers

The kit lives in the repo's .claude/. After editing it, run npm run build (pure Node — cmd / PowerShell / sh) to refresh template/, then npm version patch and npm publish. prepublishOnly rebuilds template/ on publish automatically.

Everything is cross-platform: the installer and build.js are pure Node, and the lifecycle hooks are Node scripts (node .claude/hooks/*.js) — no bash/jq/python, so the kit runs the same on macOS, Windows, and Linux.