See what every agent is doing, step by step
The Proofsource agent dashboard lists every AI visibility agent with its state, last step, next run and credits used this week. Open an agent to rename it, run it now, set its schedule and read its run history. Open any run for a step-by-step trace with timing and the credits each step used. Folders, filters, import and export keep a growing team's agents in order.
| Agent | State | Last step | Next run |
|---|---|---|---|
| Weekly lost-question fixer | Waiting on you | Draft a comparison page | Mon 9:00 |
| Citation refresh | Working | Read cited pages | Thu 6:00 |
| Competitor watch | Idle | Sum up what each site says | Tomorrow 7:00 |
 Reddit thread watch | Failed | Find threads | Retry today |
AI agent dashboard and run trace, in short
The facts, ready to quote.
- States
- Working, waiting on you, failed, idle, and published or unpublished.
- At a glance
- State, last step, next run and credits this week for each agent.
- Run trace
- Every step of a run with its status, timing and the credits it used.
- Organize
- Folders, grid or list view, filters by status, type and creator, and search.
- Portable
- Export an agent as a file and import it into another workspace.
- Durable runs
- A run keeps going through long steps and pauses, and picks up where it left off.
What is an AI agent dashboard?
Once a team runs more than a few agents, the hard question is no longer how to build one but what they are all doing. An agent dashboard answers that: which agents are working right now, which are waiting for a person, which failed and why, and what each one costs. A run trace answers the follow-up question, what exactly happened in one run.
In Proofsource, the Agents page shows every agent as a card or a row with a status badge, its type and who created it. A glance at the top sums up state, last step, next run and credits this week. Each agent has its own page: rename it inline, run it now, edit its schedule, and read its run history.
Open a run to see its trace: each step in order with what it received, what it produced, how long it took and the credits it used. When a run pauses for your approval it shows Waiting on you, so nothing that needs a person gets lost among the agents that run alone.
How it works
See every agent's state, last step, next run and credits, then open any run for a step-by-step trace with timing.
Scan the Agents page
See every agent with a status badge, type and creator, in a grid or a list.
Filter and search
Filter by status, type or creator, search by name, and group agents in folders.
Open one agent
Rename it inline, run it now, edit the schedule, and read its run history.
Read the run trace
Each step of a run with its status, timing and credits.
Share or move it
Duplicate, move, archive or export an agent, and import one from a file.
See it in the product
Screens from the Proofsource app, drawn with sample data for an example brand.
| Step | Status | Duration | Credits |
|---|---|---|---|
| Find questions that dropped | Finished | 4 s | 0 |
| For each question (3 items) | Finished | 1 s | 0 |
| Read the AI answers | Finished | 6 s | 0 |
| Named a competitor instead? | Finished | 2 s | 1 |
| Draft a comparison page | Finished | 48 s | 9 |
| Wait for your Yes | Waiting on you | 0 |
| Run | Status | Duration | Credits |
|---|---|---|---|
| Nov 2, 7:00 | Succeeded | 1 m 12 s | 6 |
| Nov 1, 7:00 | Succeeded | 58 s | 5 |
| Oct 31, 7:00 | Failed, retried | 2 m 3 s | 4 |
Every capability, live today
9 shipped capabilities behind AI agent dashboard and run trace, each in the product now.
Agents list and status badges
A grid or list of agents with search and filters by status, type and creator. Badges show working, waiting on you, failed or idle, plus published or unpublished.
Summary glance
State, last step, next run and credits this week across your agents.
Agent API endpoints
Read the agent summary, start runs and follow their progress through the Proofsource API.
Agent detail page
One agent with inline rename, run now, schedule, run history and trace.
Run trace
A step-by-step trace of one run with timing and the credits each step used.
Durable agent runs
Each run keeps going through long steps and approval pauses, including publishing, and resumes where it stopped.
Folders
Create, rename, move and delete folders for agents; deleting a folder keeps the agents inside.
Agent menu
Rename, duplicate, move to a folder, archive, delete or open in the builder.
Import and export
Export an agent definition as a file and import it to recreate the agent.
Who it is for
See what needs you
Filter to Waiting on you each morning and clear the drafts that are holding runs.
Watch credit use
Check credits this week per agent and open a trace to see which step used them.
Debug a failed run
Open the trace of a failed run to see which step stopped and what it was handed.
Keep client agents apart
Put each client's agents in a folder and export a proven agent to reuse elsewhere.
Why it matters
Nothing runs unseen
Every run has a trace with steps, timing and credits.
Approvals do not get lost
Waiting on you is its own state and filter.
Spend you can explain
Credits are shown per agent and per step.
Order at any size
Folders, filters, search and import keep many agents manageable.
Google AI OverviewsGoogle AI Mode
AI agent dashboard and run trace: Proofsource and other tools
Platforms with agents also give you a way to watch them. Here is how each describes it on its own pages.
| Capability | Proofsource | Profound | Peec AI | Writesonic |
|---|---|---|---|---|
| Central view of agent runs and usage | YesAgents page glance: state, last step, next run, credits this week. | YesCommand Center tracks agent runs, usage and feature adoption across teamSource | RelatedIn-app agent, credit-basedSource | RelatedAgentic workflow runs; full access on EnterpriseSource |
| Step-by-step trace of a single run | YesEach step with status, timing and credits. | Not listed | Not listed | Not listed |
| Run agents through an API | YesRun and summary endpoints. | YesRun Profound Agents programmatically from external toolsSource | Not listed | Not listed |
Where each of them shines
Command Center gives teams a view of agent runs, usage and adoption, and agents can be run from external tools. Compare in full
A conversational agent with memory and many connectors, with clear credit-based usage. Compare in full
Agentic workflows plus GEO content agents that create and refresh content.
Questions about AI agent dashboard and run trace
What states can an agent be in?
Working, waiting on you, failed or idle. An agent is also either published, meaning a frozen version can run, or unpublished.
What does the run trace show?
Every step of the run in order, with its status, how long it took, what it handed to the next step and the credits it used.
Can I see what my agents cost?
Yes. The glance shows credits this week, each run shows its credits, and the trace shows credits per step.
How often can I run an agent by hand?
Run now starts a run straight away, at most once an hour per agent. Scheduled runs follow the schedule you set.
What happens when a run pauses for approval?
The agent shows Waiting on you. The run keeps its place and continues when you approve, or stops when you request changes.
Can I move an agent to another workspace?
Yes. Export the agent definition as a file and import it in the other workspace to recreate it.
What happens to agents in a folder I delete?
The agents inside are kept. Only the folder is removed.
Is there an API for agents?
Yes. You can start runs, read their progress and read the agent summary through the Proofsource API.
Keep exploring
Try it on your own brand.
Proofsource is live: scan your brand and see AI agent dashboard and run trace at work with your own answers, competitors and sources.
Google AI OverviewsGoogle AI Mode