Autonomous Coding Agent¶

Autonomous Coding Agent helps VWFS teams delegate software engineering work to an AI agent that can understand a task, work with the relevant Jira issue and GitHub repository, make code changes, and report the result back to the user.

The agent is designed for development workflows: bug fixes, small features, test updates, documentation changes, refactoring, investigation, and other well-scoped engineering tasks.

How You Can Use It¶

You can start an agent in different ways, depending on how your team works.

From a chat client or app, where you can choose the agent, provide context, and continue the conversation while the task runs.
From VWFS Jira, using Jira automation rules that trigger the agent from an issue event or comment.

Important links:

Connected Tools¶

Autonomous Coding Agent integrates with the tools VWFS teams already use:

VWFS Jira for issues, task context, comments, and workflow triggers.
VWFS GitHub for repository access, branches, commits, pull requests, and code review handover.

The agent uses the team and repository access configured during onboarding. It does not replace human review; users should still review code changes before merging or deploying them.

What To Ask The Agent¶

Good tasks are specific, bounded, and easy to verify. For example:

Fix a bug described in a Jira issue.
Add a small feature with clear acceptance criteria.
Update tests after a behavior change.
Improve documentation for an existing workflow.
Investigate a failing build or test and summarize the likely cause.

Large or ambiguous requests should be split into smaller tasks so the agent can work predictably and report progress clearly.

Getting Access¶

Autonomous Coding Agent is enabled through team onboarding rather than a local installation by each user. Onboarding connects your team, Jira project, GitHub organization or repositories, and the agents your team is allowed to use.

Start with the onboarding guide to request access for your team.