Poloby Plungo
DEVOPS WORKSPACE
Assign engineering work, follow Polo’s progress, and review the results with your team.
Workspace access is protected by role permissions.
Poloby Plungo
POLO ACCOUNT
Use your workspace account to continue.
Checking your session…
Polo
Polo is your DevOps teammate. Give it a task, decide on the changes it proposes, and keep an eye on your services.
Background jobs are not running, so scheduled tasks and monitors are paused. Open Automation
RELIABILITY
Outages and alerts Polo watches, what it found and what it changed. Polo applies a fix only after someone approves it, or when a standing approval covers that exact change.
What is failing now, what Polo is doing about it, and what waits for you
How each one ended, and whether Polo’s fix is why
Where incidents come from
Uptime checks Polo runs on your endpoints. When one fails, it opens an incident and Polo investigates.
Alerts from Grafana and Slack, and the investigations they started
DECISIONS
Changes Polo wants to make and questions it has for you. Polo makes a change only after someone approves it, or when a standing approval covers that exact change.
When Polo wants to change something, the exact change shows up here first, unless a standing approval covers it. Questions Polo has for you show up here too.
Review exactly what will run, then approve or reject it
Changes you can’t approve yourself, and why
Tasks paused until someone answers
What happened to Polo’s latest changes, newest first
Each agent is a Polo teammate: what it can do, which changes need your approval and what it never does.
SETTINGS
The work Polo can take on for your team, what this workspace has connected for it, and how far each area has been tested.
Everything you asked Polo to do: what it found, the changes it proposed, and how each task ended.
WORKSPACE
Sync Polo’s calendar and track attendance requested through workspace tasks.
Connected accounts and upcoming events
Scheduled joins and verified provider status
Authorize read-only access to an existing calendar for Polo.
Choose the agent that attends meetings from your calendar and Slack, after approval.
WORKSPACE
Connect Polo’s Gmail account, route email assignments, and manage drafts and approved sends.
Connected accounts and their approved access
Authorize an existing Gmail account for one of your agents.
WORKSPACE
Review the documents and presentations Polo publishes from approved tasks.
Accounts authorized for Polo’s files
Polo requests access to the files it creates.
WORKSPACE
Give Polo scoped conversation access and approve exact messages from tasks.
Approved spaces, history, and posting permissions
For inbound mentions and direct assignments, connect your Google Chat app in Intake.
Authorize the Google identity Polo uses in approved spaces.
SETTINGS
Who belongs to this workspace, what their role lets them do, and the invitations you have sent.
Everyone in this workspace and their role
They get a link that works only for their email address
Who may approve a change also depends on the approval rules
Links you have sent and whether they were used
SETTINGS
Your account, how people sign in to this workspace, and where you are signed in.
Who you are signed in as, and what your role lets you do
Confirm the email address you use to sign in.
The password you use to sign in to Polo
This account uses its identity provider for sign-in.
Where you are signed in to this workspace
Let members sign in with Google or Microsoft, and link your own account
SETTINGS
Who must approve Polo’s changes in each environment, and the exact changes Polo may make without asking.
How many approvals a change needs, who may give them and how long a request waits
PLATFORM
Search workspace context and give Polo access to specific versions of skills.
Local semantic indexing and keyword search
Current versions, including skills
Find relevant sections and review the original version
The systems Polo works with, their status, and what Polo can reach in each.
Set up a new system, or open one to manage it.
PLATFORM
Discover tools, grant access explicitly, and review exact writes on the task.
Tool definitions and workspace grants
Use a remote Streamable HTTP endpoint.
PLATFORM
Register the protected Git branches Polo can read for assignments.
Repository identity, branch protection, and access status
Polo verifies the source and protected branch before saving access.
CONNECTIONS
See the requests and alerts coming into your workspace.
Provider, routing scope, and whether a signed delivery has arrived
Incoming events and the tasks they opened
Install Polo’s Slack app, then choose assignment and conversation scopes.
The Polo deployment administrator uses this URL for the app’s Events API subscriptions. Add Polo to the channels it should serve.
Slack Events API setupVerified project and assignee connections
Signed issue updates and the work they opened
Assign a ticket to Polo’s Jira service account to start a task.
List connections and assignment permissions
Signed assignments and the tasks they opened
Assign tasks from a ClickUp List to Polo.
SETTINGS
The AI models your agents think with. Each entry is a route to one model: its provider, the model’s name, the API key and, if you set one, its price. Each agent uses one route as its default.
Who provides each one and how its cost is counted
Choose the provider, the model and its API key, then its price if you know it
WORKSPACE
Recurring work Polo runs on a schedule, such as a weekly cost review or a daily backup check.
Uptime monitors are managed on Incidents. Go to monitors
What runs next, and the last task each one created
Polo runs it for the agent you choose, on the interval you set.
MORE
Your services, who owns them, their environments and where they deploy. Polo reads this to know what a task is about.
PLATFORM
Review workspace changes that passed offline sandbox checks, see whether Polo’s review accepted them and whether they were published or merged, and read private reports from Polo’s tasks.
Change bundles with their lifecycle, published reports, and exact downloads
SETTINGS
How much each task used its AI model, and what that cost where the model has a price.
Known cost adds up the tokens each call used at its model’s price. A call that ends without a final cost counts as not yet settled, at the most it could have cost. Calls to a model without a price, such as a local model, are counted in tokens and their cost is shown as not tracked, never as $0; each task’s token limit is what stops them. Runner, storage and cloud charges are not included.
Usage and cost for each model route
AI usage and cost for each task in this period
