/jaan-to:*
give soul to your ideas

A workflow layer for Claude Code + Codex that handles repetitive work so your team can focus on what matters.

Get Started

Read the Philosophy Available Skills

Open source. Built for Claude Code + Codex. Created by Parhum Khoshbakht

Quick Start

Install in under 60 seconds

Choose your coding runtime, run the setup steps, then call /jaan-init. Both paths unlock the same /jaan-to:* workflow skills.

See all available skills

claude
/plugin marketplace add parhumm/jaan-to
/plugin install jaan-to
# /jaan-init
# /team-sprint: Plan, build, and ship in one cycle

Then run /pm-prd-write or any /jaan-to:* skill.

bash <(curl -fsSL https://raw.githubusercontent.com/parhumm/jaan-to/main/scripts/install-codex-skillpack.sh)
# restart Codex
# /jaan-init
# /pm-prd-write New feature for Google Auth login

Use /skills, /jaan-to:*, /jaan-init, or native $skill mentions.

The Philosophy

Let AI handle the hands.
Keep humans on the mind.

AI handles

Formatting and boilerplate
Gathering context from real systems
Following templates consistently
Remembering every edge case
Producing structured outputs
AI doesn’t have a consistency problem.

Humans handle

Defining the real problem
Making trade-off decisions
Understanding customer needs
Exercising quality judgment
Knowing when to break the rules

“This is not replacing teams.
This is making teams faster.”

The System

Four building blocks.
Everything else is optional.

Skills

What to do

58 structured commands across 13 roles — PM, Data, Dev, UX, Core, QA, Detect, Release, WP, Sec, DevOps, Backend, and Frontend — with Growth planned. Each skill knows its questions, its template, and its output format.

Stacks

How your team works

Context files that describe your tech stack, team structure, and integrations. Written in Markdown. No schemas to learn.

Templates

Output format

Every skill produces from a tested template. Consistent structure, every time. Your team's voice, not the AI's.

Boundaries

Safety rules

Sandboxed file paths. Secret scanning. Preview-before-save. Explicit approval. Non-negotiable.

58 skills available now. 13 active roles. 1 more role planned.

Skills Catalog

What's available today.
What's coming next.

PM Available

/pm-prd-write — Generate PRD from initiative
/pm-story-write — Generate user stories with Given/When/Then ACs
/pm-research-about — Deep research on any topic
/pm-roadmap-add — Add prioritized items to any project roadmap
/pm-roadmap-update — Review and maintain project roadmaps
/pm-skill-discover — Detect repeated workflow patterns and suggest skills
/pm-changelog-rewrite — Changelog rewrite and maintenance workflows
/pm-sprint-plan — Assess project progress and build a prioritized sprint plan

Data Available

/data-gtm-datalayer — Generate GTM tracking code

Dev Available

/frontend-task-breakdown — Transform designs into frontend task breakdowns
/frontend-design — Create distinctive frontend component code
/frontend-scaffold — Convert designs to React/Next.js component scaffolds
/backend-task-breakdown — Transform specs into backend task breakdowns
/backend-api-contract — Generate OpenAPI 3.1 API contracts
/backend-data-model — Generate data model documentation
/backend-scaffold — Generate production-ready backend code with routes and services
/dev-project-assemble — Wire backend + frontend scaffolds into runnable project
/backend-service-implement — Generate full service implementations with business logic
/dev-output-integrate — Copy generated outputs into operational project locations with config merging
/dev-verify — Combined build and runtime verification for integrated projects
/backend-pr-review — Multi-stack backend PR review for security, performance, and code quality
/frontend-story-generate — Generate CSF3 Storybook stories with CVA variant detection
/frontend-visual-verify — Visually verify components using Playwright MCP
/frontend-component-fix — Diagnose and fix UI bugs with output-only safety model

UX Available

/ux-heatmap-analyze — Analyze heatmap CSV exports and screenshots
/ux-microcopy-write — Generate multi-language microcopy packs
/ux-research-synthesize — Synthesize UX research into actionable insights
/ux-flowchart-generate — Generate GitHub-renderable Mermaid flowcharts

Core Available

/docs-create — Create documentation
/docs-update — Audit and maintain documentation
/learn-add — Add lesson to skill's LEARN.md
/skill-create — Create new skill with wizard
/skill-update — Update existing skill
/jaan-init — Initialize Jaan.to in a project
/jaan-issue-report — Report bugs, feature requests, or issues to GitHub
/team-ship — Assemble role-based AI teammates to ship ideas from concept to production
/team-sprint — Full development sprint cycle orchestrator from planning to PR

QA Available

/qa-test-cases — Generate BDD/Gherkin test cases from acceptance criteria
/qa-test-generate — Produce runnable Vitest unit tests and Playwright E2E specs from BDD test cases
/qa-test-run — Execute tests, diagnose failures, auto-fix, and generate coverage reports
/qa-issue-report — Report issues to any GitHub/GitLab repository
/qa-test-mutate — Run mutation testing to validate test suite quality
/qa-tdd-orchestrate — Orchestrate RED/GREEN/REFACTOR TDD cycle
/qa-contract-validate — Validate API contracts through 4-tool pipeline
/qa-quality-gate — Compute composite quality score from upstream outputs
/qa-issue-validate — Validate GitHub/GitLab issues against codebase

Release Available

/release-iterate-changelog — Generate changelogs with user impact notes from git history

Detect Available

/detect-dev — Engineering audit with OpenSSF-style scoring
/detect-design — Design system detection with drift analysis
/detect-writing — Writing system extraction with tone analysis
/detect-product — Product reality extraction with evidence model
/detect-ux — UX audit with route extraction and heuristics
/detect-pack — Consolidate detect outputs into knowledge index

WP Available

/wp-pr-review — Review WordPress plugin PRs for security and standards

Sec Available

/sec-audit-remediate — Generate targeted security fixes from SARIF findings

DevOps Available

/devops-infra-scaffold — Generate CI/CD workflows, Dockerfiles, and deployment configs
/devops-deploy-activate — Activate deployment pipeline with secrets, platform provisioning, and verification

MCP Available

Context7 — On-demand library documentation via /dev-docs-fetch
Storybook — Component conventions and story patterns
shadcn — Component prop types and design tokens
Playwright — Visual verification and accessibility testing

20 more connectors planned — GA4, GitLab, Jira, Figma, and more

Growth Coming Soon

SEO and content workflows — coming soon

Team Sprint

One command.
A full sprint cycle, start to finish.

Run a complete development sprint — from project assessment through planning, team execution, and PR delivery — with a single command. Human approval at every phase gate.

/team-sprint → /pm-sprint-plan → /team-ship --track sprint

PM UX Backend Frontend QA DevOps Security

1 Pre-flight Create cycle branch, verify plugin, security check

2 Plan Invoke /pm-sprint-plan — assess state, build queue

hard stop — approval required

3 Execute Invoke /team-ship — Agent Teams build in parallel

4 Post-cycle Mark ROADMAP, write gap report, create PR

$ /team-sprint 5 --focus code --tasks "auth,payments"

> Pre-flight: cycle branch created, plugin verified
> Planning: /pm-sprint-plan assessing project state...
> Sprint plan ready. [Awaiting your approval...]

> Executing: /team-ship --track sprint
> 7 teammates building in parallel...
> Post-cycle: gap report written, PR created.

Sprint Plan artifact · /pm-sprint-plan Prioritized execution queue with task breakdown

Gap Report artifact · post-cycle What was completed, what remains, what to do next

Pull Request deliverable · dev branch Production-ready code with changelog and tests

The Vision

One person can ship like a team.
One team can ship like a company.

Now

58 production skills. Learning system that improves from feedback. Two-phase workflow with human approval. Open source, MIT licensed.

60+ skills across 14+ product roles. MCP connectors to real systems — Figma, Jira, GA4, Sentry, Clarity, and more.

Ahead

Multi-agent compatibility. Marketplace distribution for community skills. The smallest AI company that can ship.

Read: "The Smallest AI Company That Can Ship" The story behind Jaan.to, on Medium.

Give your product work a soul.

Install from Section 1, then use docs + source to customize workflows for your team.

Get Started on GitHub Read the Docs

Created by Parhum Khoshbakht

LinkedIn Medium GitHub

/jaan-to:* give soul to your ideas