# How you see what they did See everything. Miss nothing. A real-time activity feed shows who is working on what right now. The step-by-step inspector traces every tool call, decision, and reasoning chain. Cost tracking breaks down spend per message, per employee, and per team. Walk through your live 3D office to see your workforce at a glance. The observability system gives you complete visibility: a real-time activity feed, a step-by-step execution inspector, per-message cost tracking, and a live 3D office view. Every action, tool call, decision, and cost is recorded and visible. Transparency is how you build confidence that your AI workforce is doing the right things, the right way, at the right cost. It works in layers, so you go as shallow or as deep as the question needs. The feed answers what is happening right now. Open any entry and the inspector answers what exactly happened: the original request, what the employee planned, which tools it called with which inputs, what came back, and what it decided from that. The work journal is the same story told as a narrative, useful when you want the week rather than the run. Cost is attached to the work rather than delivered as a shock at the end of the month. Every message shows what it cost in credits, every employee carries a running total, and every team rolls up. When one employee costs noticeably more than another for similar work, the inspector shows where the credits actually went, which is usually a heavier model than the job needed or a tool being called more times than it should be. You act on the data instead of guessing. You can also put a ceiling on it. A spending limit caps how many credits your workforce may use per hour, day, week or month, so a runaway task or an unusually busy day cannot turn into a surprise. Work already in flight is never cut off halfway, and when you do hit the limit, one click releases the next period's allowance. It is a brake you control, not a switch that strands an employee mid task. Knowing that work happened is not the same as knowing it was any good, so there is a reviewer for that too. Turn it on and each real deliverable an employee produces is scored against a standard, with the verdict shown in the inspector next to the run that produced it, while everyday chat is left alone so you only grade actual output. Team dashboards cover the level above: each team gets its own view of health, current sprint, work in flight, momentum, objectives and KPI progress. And when you need the records outside Sistava, every history screen has a download control that exports the current filtered view as Markdown, CSV or JSON. ## At a Glance - **Real-time** Activity feed - **Step-by-step** Execution traces - **Per-action** Cost tracking ## Benefits ### Full visibility A live, continuously updating timeline of everything happening across your AI workforce. Filter by employee, team, or action type. Delegation chains and multi-agent coordination are visible in real time. ### Execution inspector Click any item and see the full story: what was planned, which tools were called, what each returned, what the agent decided, and why. Debug AI work the same way you debug human decisions. ### Cost control Every message shows its credit cost. Every employee shows a running total. Every team shows aggregate spending. Compare cost efficiency across employees and optimize model assignments based on real data. ## How It Works **See exactly what every AI agent is doing, what it cost, and what happened at each step, in real time.** The real-time activity feed shows every action taken by every AI employee on your team as it happens. You do not have to ask for a status update. The feed streams actions, tool calls, decisions, and completions continuously. When something goes wrong, you know immediately, not after the fact. The step-by-step execution inspector lets you drill into any task and see exactly what the agent did: which tools it used, what inputs it passed, what outputs it received, and how long each step took. Cost tracking is built in at the action level, so you know precisely what each task costs in compute and API spend. Notifications alert you to completions, failures, and anything requiring your attention. None of this asks you to sit and watch a screen. Notifications reach you in the app, by email and in Slack when a task finishes, an approval is waiting, an employee is blocked or your credit balance is running low, and you can reply straight from the notification with the context already attached. Spending limits and the quality reviewer work the same way, quietly in the background, surfacing only when something genuinely needs a decision from you. The dashboards are there for when you want to look, not a chore you have to keep up with. ## Use Cases ### See exactly what your employee did and why The activity feed shows every action taken, every tool used, and every decision made in order. No guessing what happened. Full transparency into every task. ### Inspect any task step by step Open the inspector on any completed or in-progress task to see inputs, outputs, tool calls, and timing at each step. Debug issues and understand behavior precisely. ### Track costs across your entire workforce See token usage and cost per employee, per task, and per time period. Know exactly what you're spending and which employees are delivering the most value per dollar. ### Watch your team work in the 3D office The 3D office view shows all your AI employees as live avatars. See who's active, what they're working on, and how the team is moving in real time. ## FAQ ### How do I see what my AI employee actually did and what it cost? Open the activity feed in Sistava. It is a live timeline of every action across your workforce, and each entry carries the credit cost of the work behind it. Click an entry and the inspector opens, showing what the employee planned, which tools it used, what each returned and what it decided. Filter by employee, team, action type or date range when you are hunting for one specific thing. ### Can I find out why an employee made a particular decision? Yes. The inspector shows the reasoning behind a run and not just its outcome: the request that started it, the plan, every tool call with its inputs and results, and the conclusion the employee reached from them. When something went wrong, this is where you see the exact point it went wrong. When it went right, it confirms the process was sound rather than lucky. ### Can I cap how much my AI workforce is allowed to spend? Yes. A spending limit sets a ceiling on credits per hour, day, week or month across your whole workforce. Work already running is never interrupted when the ceiling is reached, and one click releases the next period's allowance if you want to carry on. Alongside it you can see spend per message, per employee and per team, so you know what is actually pushing against the limit. ### How do I know the work is actually good and not just finished? Sistava can score it for you. A quality reviewer grades each deliverable an employee produces against a standard and shows the verdict in the activity inspector beside the run that made it. It stays off until you switch it on, and it only looks at real output such as a document, an email or a completed task, so ordinary conversation is never graded. You get a quality signal per piece of work instead of a general impression. ### Do I have to watch a dashboard all day to stay on top of this? No. Notifications come to you in the app, by email and in Slack when a task completes, an approval is waiting, an employee is blocked or your credits run low. You choose which kinds of event are worth interrupting you and which can sit quietly in the feed. You can reply straight from a notification and the context travels with it, so you never have to go and find the conversation first. ### Can I see how a whole team is doing rather than one employee at a time? Yes. Every team has its own dashboard showing team health, the current sprint, work in flight, momentum, objectives and KPI progress, built from data Sistava already holds with nothing for you to connect. The activity feed also filters by team, so you can watch a department the way you watch one employee. Handoffs between employees show up too, so you can follow a task from the leader to the specialist and back. ### Can I get my records out of Sistava for an audit? Yes. Chat threads, the task board, activity logs and Drive files each have their own download control that exports the current, filtered view as Markdown, CSV or JSON. That means you can pull exactly the slice you need, for one employee or one date range, rather than a whole database dump. What you export matches what you were looking at on screen. ### Does the history disappear after a while? No. Activity history is kept for your workspace rather than trimmed to a rolling window, so you can scroll back to your first day and open the trace for a run from months ago. Execution traces are kept alongside it, with no shorter separate limit of their own. That is what makes the feed usable as a record you can answer questions from later, not just a live view. ## Full Visibility at Any Scale Observability degrades on most platforms as team size grows. On Sista, the activity feed, cost tracking, and execution inspector are scoped per agent but aggregated across your workforce. You can zoom into one agent or view the entire team at once. Nothing gets lost in the noise. The live 3D office view gives your team a spatial representation of agent activity. It is not just a novelty: it surfaces which agents are active, idle, or blocked at a glance. Combined with notifications and cost alerts, you have everything you need to manage a large AI workforce without constant manual checks. ## How It Works 1. **Watch the activity feed**: Every agent action streams in real time. Filter by agent, task, or tool to find exactly what you need. 2. **Inspect any execution**: Drill into any completed or in-progress task to see every step, every tool call, and every decision the agent made. 3. **Track costs per action**: Cost is tracked at the individual action level. Set alerts and review spend by agent, task type, or time period. > The execution inspector showed us our AI agent was calling the same API three times per task. We fixed it in ten minutes and cut our costs by 40 percent. > > Sofia A., CTO ยท SaaS company ## Features in Monitoring - [Activity Timeline](/en/features/observability/activity_timeline): See every piece of work your AI employees do as it happens: chat replies, scheduled runs, tool calls, delegations, and finished deliverables, all logged as activities you can open and inspect. It runs automatically for every employee with no setup, and a live feed in the workspace sidebar shows actions as they occur so you never have to wonder what is happening right now. - [Action Inspector](/en/features/observability/activity_inspection): Click into any single activity and see exactly what your AI employee did: the model call pipeline, every tool it used with inputs and outputs, cost broken down by category, a duration waterfall, and the plan it followed. Available from the Activity tab, from Inspect on any chat message, or from the sidebar timeline. - [See What Your Employee Is Told](/en/features/observability/prompt_inspection): Open any activity in the Activity Inspector and read the exact system prompt your AI employee received for that run: every instruction, every tool definition, and the manifest of which skills and duties were loaded and why. Nothing paraphrased, nothing summarized, the same text the model saw. - [Credit Usage History](/en/features/observability/credit_usage_history): See exactly where your monthly credits go: split spend from direct employee requests and automation or system work, then inspect every charged system operation, including AI work, tools, Officer checks, knowledge, voice, meetings, video, and email. A burn chart shows the trend over time, and a separate grants table lists every bonus credit you received from promos, referrals, or purchases, with source and date. - [Rate Limits](/en/features/observability/spend_limit): Rate Limits sets a ceiling on how many credits your workforce can spend per hour, day, week, or month, so a runaway task or a busy day never turns into a surprise bill. - [Team Dashboards](/en/features/observability/team_dashboards): Every team gets its own dashboard the moment you open it: team health, the current sprint, work in flight, momentum, objectives, and KPI progress, all built from data the platform already has. On the Custom plan, connect data sources and the same dashboard adds Revenue & Pipeline, Demand & Conversion, and Reach & Engagement cards for that team. - [Data Export](/en/features/observability/data_export): Every workspace surface that shows your history, chat threads, the task board, activity logs, and Drive files, has its own download control that exports the current, filtered view as Markdown, CSV, or JSON. Use it to pull records for an audit, share a run outside the app, or keep an offline copy of what an employee produced. - [AI Evals](/en/features/observability/quality_evaluation): AI Evals reviews selected employee work after it is produced. A quality reviewer scores each deliverable against a standard and shows the verdict in the activity inspector. Everyday chat is left alone, so you only grade real output. - [3D Office View](/en/features/observability/office_3d): A workforce of AI employees is easy to lose track of as a list of chat threads. The 3D Office View turns it into a place: every employee is a character at a desk, hot-desk, or the lounge, moving between them live as their status changes, so you can see your whole company working at a glance instead of clicking through tabs. - [In-App Notifications](/en/features/observability/in_app_notifications): See the moment an Employee finishes work, needs a decision from you, or your credit balance runs low, through a bell icon in your workspace that always fires. Reply straight from the notification drawer for Employee updates and comment threads, and control which events also reach your email through a short list of categories. - [Reply to Notifications](/en/features/observability/notification_reply): Reply straight from the notification drawer, whether it is an update from an Employee or a comment on your work, and your response goes back with the full context of what it is replying to. No need to open the chat or dig up the artifact first. - [Email Notifications](/en/features/observability/email_notifications): Sistava emails you when an employee finishes work, needs your approval, hits a blocker, or your credit balance runs low, so nothing important gets stuck waiting for you to open the app. You choose which categories of email you want; account security and billing mail stay on because they carry information you need regardless of your other preferences. - [Email Digest Notifications](/en/features/observability/email_digest_notifications): A rolled-up email that summarizes what your AI employees did over a set period, instead of a separate message for every event. Today, notification email in Sistava sends per event and per category (employee updates, task completion, credit alerts, and so on); a digest option that batches those into one scheduled email is planned but not built yet. ## Explore - [Every feature](/en/features) - [Browse AI employees](/en/agents) - [Pricing](/en/pricing)