Code checkout
Repository access stays with the worker you operate.
Run AI coding on your own infrastructure.
The worker is where the sensitive work happens: checkout, context assembly, AI execution, validation, model calls, and review preparation.
Works with
Jira GitHub GitLab M monday.dev Linear Azure Boards Azure Repos
Jira GitHub GitLab M monday.dev Linear Azure Boards Azure Repos
Jira GitHub GitLab M monday.dev Linear Azure Boards Azure Repos
Jira GitHub GitLab M monday.dev Linear Azure Boards Azure Repos
Context assembly
Ticket context and repository rules are assembled inside your environment.
Model calls
Use Codex CLI, Claude Code CLI, OpenAI-compatible endpoints, Vertex AI, AWS Bedrock, or Azure Foundry.
Local validation
Run setup, tests, and custom validation commands where dependencies already exist.
Review prep
Branches, commits, PRs, and MRs are prepared for your normal review path.
Run inside your boundary
The sensitive parts of AI coding happen on your worker: code checkout, context assembly, model calls, tests, and PR/MR preparation.
Code stays controlled
Repository access, dependencies, credentials, and validation commands stay in the infrastructure your team operates.
Approved providers only
Use Codex CLI, Claude Code CLI, OpenAI-compatible endpoints, Vertex AI, AWS Bedrock, or Azure Foundry according to your policy.
Security can explain it
Teams can answer where the run executed, which tools ran, what provider was called, and what checks completed before review.
Worker responsibilities
Keep checkout, AI execution, validation, and review preparation in one controlled place inside your environment.
Code checkout
The worker checks out repositories using credentials and access patterns you control.
AI execution
The agent run happens where your tooling, network rules, and policies can be enforced.
Validation
Setup, test, lint, and custom commands run before the PR or MR reaches review.
Review request prep
The branch and review request return to GitHub, GitLab, or Azure Repos.
Avoid building your coding workflow around one AI vendor
CLI agents
Use approved agent tools such as Codex CLI or Claude Code CLI.
Compatible endpoints
Use OpenAI-compatible endpoints when they support the required tooling.
Cloud platforms
Route through approved platforms such as Vertex AI, AWS Bedrock, or Azure Foundry.
Same workflow
Change AI provider, issue system, or code host without changing how tickets become PRs or MRs.
Runs with
CX Codex Claude Vertex AI AWS Bedrock AZ Azure Foundry API OpenAI-compatible
CX Codex Claude Vertex AI AWS Bedrock AZ Azure Foundry API OpenAI-compatible
CX Codex Claude Vertex AI AWS Bedrock AZ Azure Foundry API OpenAI-compatible
CX Codex Claude Vertex AI AWS Bedrock AZ Azure Foundry API OpenAI-compatible
See what else MergeLoom can do.
Connect more of your stack, improve context, validate output, and keep audit evidence across every AI coding run.
Try one controlled AI coding workflow.
Start with one tracker, one repository, and one validation path before rolling AI coding across the team.