Set up Bunny
Create an AI Employee and configure its Redmine account, role, model, and instructions.
Bunny Redmine by Appar
Your AI Agent team inside Redmine
AI Agents work directly in Redmine to plan projects, organize issues, execute tasks, and track progress.
Connect privately deployed or cloud AI inference models.
OpenAI GPTClaudeGeminiGrokDeepSeekQwenLocal LLM
MEET YOUR AI TEAMMATE
A conversation that turns into real project work.
Give Bunny a requirement. Watch it organize the work directly in Redmine.
Illustrative product walkthrough
Your AI teammate, right inside Redmine.Start with a message and the documents you already have.Watch the tools run. Projects, issues, and documentation stay together.Bunny prepares the work. You keep the final decision.
More than a chatbot
Bunny doesn’t just answer questions. It gets work done in Redmine.
Bunny has the tools to carry out Redmine tasks and scheduled work. It can also connect to other enterprise systems through MCP.
Give it the tools to work
Create projects and issues, assign owners, update status, manage Wiki, News, and Forums, and use browser tools.
Let it plan the next task
Bunny can decide what to do next and create or adjust one-time and recurring schedules.
Verify results and keep records
Check outcomes, attach evidence, and record the process. Request confirmation when your rules require it.
An enterprise AI hub
More than project management. Redmine becomes your enterprise AI hub.
Bunny connects department data, coordinates AI Agents, and completes tasks in Redmine.
CROSS-DEPARTMENT COMMUNICATIONBunny connects department data and completes tasks in Redmine.
Illustrative product walkthrough
Agents use MCP Server to work with permitted Redmine context.Bunny uses MCP Client to request context from approved external systems.Project work and external information come together in one AI workflow.Track progress and outcomes with reports, documents, and source references in Redmine.
Comprehensive Redmine tool calls
Create and manage projects, subprojects, issues, versions, assignments, Wiki, News, Forums, and attachments within account permissions.
Your Bunny AI team
Deploy AI Employees for different roles, each with its own prompt, private skills, notebook, model, tools, schedules, and monthly budget.
Connect department systems
Use MCP Client to retrieve data and request actions from department systems. MCP Server exposes permitted Redmine tools to other agents, enabling cross-agent collaboration.
AI Employees by role
AI teammates for every stage of enterprise work
Project & Program Manager
Monitor projects, dependencies, workload, overdue work, and delivery risk. Prepare status reports and turn management decisions into assigned Redmine issues.
See more →Product Manager & Business Analyst
Turn research, customer feedback, meeting notes, and documents into requirements, user stories, acceptance criteria, roadmaps, and release notes.
See more →Software Developer
Delegate permissioned issues to Cursor, Claude or Codex, run planning, coding, tests, and reviews in parallel, then return implementation evidence to Redmine.
See more →QA & Quality Manager
Generate test cases, use Bunny’s browser tools, annotate screenshots, retain test evidence, and prepare issues for review.
See more →IT Operations & Service Desk
Triage incidents and service requests, classify and assign work, maintain operational knowledge, follow remediation, and keep stakeholders updated.
Customer Support & Feedback
Cluster incoming feedback, link duplicate reports, create actionable issues, route them to the right owners, and follow each request through resolution.
See more →Full feature set
Bunny Redmine’s execution and integration capabilities
Native Redmine AI Agents
Create and update projects, subprojects, versions, issues, assignments, Wiki pages, News, Forums, and attachments.
Autonomous work and scheduling
AI decides what to do next and when to run it, creating or adjusting one-time and recurring schedules.
Redmine MCP Server
Connect other AI clients and agents, such as Codex and Claude. Redmine tools are exposed according to each user’s account permissions.
MCP Client and enterprise AI hub
Let Bunny request data or actions from approved enterprise systems and coordinate cross-system work from Redmine.
Skills, prompts, and notebook memory
Give every employee a private AI twin with role-specific instructions, reusable skills, and knowledge that improves with use.
Reports, dashboards, and notifications
Generate recurring project reports, maintain AI-managed dashboards, publish Redmine News, and notify teams by approved channels.
Use cases
How our customers use Bunny Redmine
Automated project management and tracking
Bunny breaks down project work, assigns issues, follows up on progress, and writes activity records and weekly reports.
Let coding agents resolve Redmine issues
Cursor, Claude, or Codex can read permitted issues and images through MCP, carry out tasks, and post comments, screenshots, and validation results. They can update issue status and assign work to the right person.
Run QA with built-in browser tools
Bunny uses its own browser tools to run tests and retain the testing process. It can annotate screenshots and add written explanations to document findings on the issue.
Spot project delays before they grow
Bunny checks schedules and stalled work, identifies potential risks, follows up with the responsible people, and records their updates.
Turn Redmine into an AI-powered knowledge system
Search and organize permitted Wiki pages, files, issues, News, and Forums. Bunny connects related information and updates documentation as project work changes.
How security works
Complete permission controls for your AI Agents
Each AI Employee is bound to a Redmine account. Project membership and roles determine what it can read and change; tool controls further limit which actions it can perform.
ADMINISTRATOR CONTROLSClear permission policies. Every execution recorded.
AI EmployeeProject assistant
AI can access only what its Redmine account can access.
Only the tools you allow are available to this Employee.
At the monthly limit, the next job stops.
See what ran, what was blocked, and why.
Bound Redmine accountProject manager · pm-teamSame project permissions as this account
Permitted projectAccessible
Restricted projectNot accessible
Tool allowlist
Only selected tools are available to this AI Employee.
create_issueAccessiblemanage_wikiAccessiblecreate_projectNot accessibleHard limit$50 USD
Budget reached. Execution stopped.
Start next scheduled jobBudget stop
Execution logs.json
ActionResult
Read permitted issueCompleted
Read restricted projectPermission denied
Start next scheduled jobBudget stop
View example JSON record
{
"example": true,
"account": "pm-team",
"tool": "get_issue_content",
"status": "completed",
"monthly_budget_usd": 50,
"next_execution": "budget_exceeded"
}Configured by system administrators
Illustrative product walkthrough
Bunny inherits the bound account's access. Restricted projects stay restricted.Administrators choose the tools each AI Employee can use.A monthly hard limit stops execution when the configured budget is reached.Review the outcome and execution context in a structured JSON record.
Account-bound permissions
Every AI Employee maps to a Redmine account. If that account cannot see or change something, the AI cannot either.
Administrator-only governance
Only system administrators can configure other users’ AI Employees, models, tools, prompts, skills, notebooks, and schedules.
Tool-level allowlists
Give each AI Employee only the tools required for its role. Unnecessary or sensitive actions stay unavailable.
Approval rules and emergency stop
Prompts define which actions can take effect immediately and where human confirmation is required. Administrators can stop an execution at any time.
Complete structured audit trail
Context, tool calls, changes, results, token use, cost, errors, and timing are retained as JSON for review and auditing.
Your models and API keys
Use administrator-approved cloud or local models with customer-provided API keys and no Bunny markup on model token charges.
Choose your path
Create a new Redmine environment or upgrade your existing one
Appar plans and deploys the solution to match your needs.
Bring AI into the work
Let AI accelerate your business workflows
Tell us how your Redmine is deployed today. We’ll recommend an approach for your needs.
FAQ
Frequently asked questions
What is Bunny Redmine?
Bunny Redmine adds governed AI agents to Redmine. They can create and update work, run schedules, use approved tools, coordinate external agents through MCP, and return results to Redmine for review.
Do we need a new Redmine installation?
No. Appar can add the Bunny AI plugin to an existing Redmine or provide a new managed environment with Bunny AI built in. Contact Appar to plan either path.
Can Bunny access data outside a user’s Redmine permissions?
No. Every AI Employee is bound to a Redmine account and inherits its permissions. If that account cannot access a project or issue, the AI cannot access it either.
Can we use our own AI model or run one locally?
Yes. You provide the API key and an administrator assigns an approved cloud or local model to each AI Employee. Bunny does not add a markup to model token charges.
What is the difference between the MCP Server and MCP Client features?
The MCP Server lets approved clients such as Cursor or Claude work with permitted Redmine data and tools. The MCP Client lets Bunny call approved external systems, making Redmine a hub for cross-system AI work.
Can Bunny work autonomously without human approval?
Administrators define the boundary. Approved actions can take effect immediately, while prompts can require human confirmation at specific steps. An administrator can stop an execution at any time.
How are AI cost and activity controlled?
Administrators can set monthly company and per-Employee hard limits. Execution context, tool calls, changes, results, token use, cost, errors, and timing are retained as structured JSON for auditing.
What happens when the annual subscription ends?
The installed plugin continues to work. Annual renewal provides product updates and priority issue handling; without renewal, you can keep using the current version but will not receive updates.