What AI models do you support?

We support Claude Haiku 4.5 for fast, cost-effective work — recommended for most use cases. For complex reasoning and architectural decisions, we offer Claude Opus 4.5 and Claude Opus 4.6, the latest and most capable model. You can switch models per project at any time.

What integrations do you support?

Synlets integrates with Jira and Asana for ticket management, GitHub and GitLab for repositories and pull requests, and Confluence and Notion for knowledge bases. Connect your existing tools in minutes — no custom setup required.

How does billing work?

You purchase ACU (AI Compute Unit) credits that are consumed per task. Each agent iteration uses a set amount of ACUs based on the model and complexity. You only pay for what you use — no idle costs, no per-seat fees.

Do you store my private repository code?

No. Your codebase is retrieved directly from GitHub or GitLab at agent runtime and is cleaned immediately after each task completes. We do not store repository contents in our database. Agent context and code analysis happen in-memory during execution and are discarded once the task is done.

Security is built into every layer. All data is encrypted in transit and at rest. We use isolated execution environments for each agent task. Repository access is scoped to the minimum required permissions, and you can revoke access at any time from your GitHub or GitLab settings.

Can I try it before committing?

Yes. Our free tier includes ACU credits so you can test the full platform — create a project, connect a repo, and let an agent implement a real ticket. No credit card required to get started.

What happens if the AI gets something wrong?

Every change the AI makes goes through a pull request — nothing is merged without your approval. Our PR review agents also catch issues automatically. If you leave feedback on a PR, the agent reads it, iterates on the code, and pushes fixes. It's a feedback loop, not a black box.

Quick Start Guide

Get started with Synlets in 7 steps

From sign-up to your first AI-generated pull request in minutes. Connect your tools, assign a ticket, and let your AI teammate handle the rest.

Your path to autonomous development

Each step takes a few minutes. Most teams are up and running within an hour.

Create your account

Sign up for Synlets in under a minute. No credit card required — start with a free trial to see how AI agents work with your codebase.

Email + password or OAuth sign-up

Create your organization

Invite team members anytime

Connect your repositories

Link your GitHub or GitLab account so AI agents can read your code, create branches, and open pull requests — just like a human teammate.

GitHub & GitLab supported

Select which repos agents can access

Fine-grained permission controls

GitHub

GitLab

Connect your ticket management

Link Jira or Asana so agents can pick up tickets, update statuses, and comment on progress. Your existing workflow stays exactly the same.

Native Jira & Asana integration

Agents read ticket descriptions & context

Status updates happen automatically

Jira

Asana

Create a project

Projects are the central piece of Synlets. Compose your repositories, ticket management, and knowledge sources into a single workspace that gives agents all the context they need.

Link repos + ticket board + knowledge base

One project per product or team

Add Confluence, Notion, or file uploads for context

Assign or generate tickets

Assign existing Jira or Asana tickets to an AI agent — or use the built-in chat to generate new, well-structured tickets and assign them to an agent or a human teammate.

Assign any existing ticket to AI

Generate new tickets from natural language

Assign generated tickets to agents or humans

Existing tickets → assign to agent → get a PR

Chat → generate ticket → assign to agent or human

Bonus: PR review agent

Optionally configure a PR review agent to review pull requests from both AI agents and human engineers. Instead of leaving inline comments, it prepares a working solution on a child branch of your PR — accept it, reject it, or ask for changes. Like having a senior engineer reviewing every PR.

Reviews AI and human PRs alike

Pushes fixes to a child branch, not just comments

Pairs with project agents for an automated feedback loop

Accept, reject, or reiterate — just like a human reviewer

Aligns every PR with your docs & coding standards

Bonus: insight reports

Generate thorough AI-powered reports on any project — code quality, architecture health, technical debt, and more. Great for non-technical stakeholders who want to understand the state of the product without reading code. Works on any codebase, even before agents touch it.

Assess any codebase on demand

Non-technical stakeholders can generate reports

Export as PDF for leadership & clients

Understand your product's health without reading code

Generate anytime — takes seconds, not hours

Tips for the best results

Write clear ticket descriptions

The more context in your ticket — acceptance criteria, expected behavior, relevant files — the better the agent's output.

Upload your documentation

Connect Confluence, Notion, or upload files to the knowledge base. Agents reference these when making implementation decisions.

Start with smaller tickets

Begin with well-scoped tasks like bug fixes or feature additions. As you build confidence, assign larger tickets.

Enable the PR review agent early

The review agent catches issues before you look at the PR. It also teaches project agents your team's standards over time.

Ready to put your backlog on autopilot?

Connect your repos and ticket board. Let AI agents handle the implementation, code review, and iteration.

No credit card required

Free credits to start

Your code stays yours