The next phase of coding with LLMs

One CLI to run Claude Code through acceptance criteria, quality gates, and automated iteration—until tests pass.

Project credentials

What is Sequant

How It Works

Workflow automation for Claude Code. Sequant turns GitHub issues into merge-ready pull requests through structured phases and quality gates.

Claude Code is remarkably capable. But capability without process means results may vary — vague issues produce vague implementations, there's no formal review, and context drifts over long sessions. Sequant provides the discipline: a repeatable pipeline where every issue follows the same lifecycle.

Phase Isolation

Each phase runs in a fresh session inside an isolated worktree. No context drift — code never touches main until gates pass.

8 Quality Gates

Security, types, scope, tests, build regression, and anti-patterns — checked automatically on every PR.

CLI Automation

Batch issues, chain dependencies, and resume failed runs. Process your backlog overnight with cost-efficient sub-agents.

Issue #123
Spec Plan
Reads issue, lints AC, posts plan to GitHub
Isolated Worktree
Exec Build
Implement, commit, push, create PR
QA Review
Type safety, security, scope, tests
Loop Auto-fix
Up to 3 retries
Merge PR lands on main

What runs underneath


Skills

18 slash commands that tell Claude Code what to do in each phase

Hooks

Pre-tool blocks secrets and unsafe commands. Post-tool formats and observes quality.

State

Tracks phase progress, metrics, and run logs — updated at every phase boundary

GitHub Comments

Plans, verdicts, and phase markers that carry context between sessions

A Smarter Way to Code

2 Ways to Solve Issues

( 01 ) Automation
Sequant Run

Point it at an issue. Walk away. Come back to a finished PR — spec'd, built, tested, and QA'd.

Sequant Run illustration
or
Interactive ( 02 )
Full Solve

Stay in the loop. Run each phase — /spec, /exec, /test, /qa — with full control.

Fullsolve illustration

Skills

Commands

Run any skill by name in conversation, as a /slash command, or via npx sequant run. Not sure where to start? Try /solve.

Core Workflow Automation Testing & Verification Analysis & Utilities

Quality Loop

Automated Iteration

When tests fail, Sequant doesn't stop. It analyzes the failure, fixes the code, and retests — automatically. You review working code, not broken builds.

Enable it:

Or interactively: /loop 123. Built into /fullsolve by default.

Control retries with --max-iterations 5. Gate chains with --qa-gate.

3 Max retries
for test failures
2 Max retries
for QA issues
0 Manual
intervention
1
Build /exec implements the feature in an isolated worktree
2
Test Run tests, type checks, and quality gates automatically
3
Analyze Parse failures — type errors, missing AC, security issues
4
Fix Apply targeted changes to address each identified issue
Repeat Loop until all gates pass — or hand off for manual review

Quick Start

Zero to First PR

Install ( 01 )

Install the package, then scaffold your project.

Requires Claude Code and GitHub CLI (gh auth login)

Assess ( 02 )

Point it at one or many GitHub issues.

Analyzes your issues and recommends the right workflow — including npx sequant run flags like --chain, --sequential, and --quality-loop. Run reference →

Build ( 03 )

Full pipeline, headless, or step by step.

Interactive:

Spec → exec → test → QA with auto-fix iterations.

or
Headless:

Walk away. Same pipeline, no conversation needed.

Or run individual skills:

Skills also work without slash commands — just describe what you need. Cheat sheet →

Ship ( 04 )

QA passes. Merge the PR.


Quickstart guide →   Full docs →

FAQ

Common Questions

What is Sequant?

A CLI tool that adds structured workflows to Claude Code. It orchestrates your AI coding sessions through defined phases — spec, exec, test, qa — with automated quality gates. Think of it as a workflow engine between your GitHub issues and Claude Code.

How is this different from using Claude Code directly?

Claude Code is freeform conversation. Sequant adds structure: acceptance criteria extracted from issues, isolated git worktrees for safe experimentation, automated test iteration, and quality gates that verify code before merge. You get the speed of AI with the confidence of a defined process.

What are the prerequisites?

Claude Code (requires a Claude subscription), GitHub CLI (gh auth login), Git, and Node.js 18+. Optional: Chrome DevTools MCP server for browser-based testing, Semgrep for enhanced security scanning.

Does it work with existing projects?

Yes. Run npm i sequant && npx sequant init in any git repository. It auto-detects your stack (Node.js, Python, Rust, Go, Next.js, Astro, and more) and configures accordingly. The worktree-based workflow works with any git repository.

What happens if the quality loop can't fix something?

The loop has bounded retries — 3 for test failures, 2 for QA issues. If a fix requires architectural decisions or ambiguous requirements, it stops and hands off for manual review. You're never stuck in an infinite loop.

Is my code sent anywhere?

Sequant collects zero telemetry, analytics, or user data. The only external communication is GitHub CLI operations you initiate (reading issues, creating PRs). Your code stays local.

Does it work with other AI models?

Sequant is built specifically for Claude Code. The workflow concepts are model-agnostic, but the current implementation requires Claude Code as the underlying AI engine.

Can I customize the workflow?

Extensively. Add project-specific coding standards via a constitution file, create custom slash commands, configure phase selection (skip spec, skip test), set custom base branches, and adjust loop iteration limits. Customizations live in .claude/.local/ and are never overwritten by updates.

Changelog

Latest Releases