AgentOps Case Control Room

Turn AI-agent work into a UiPath-governed case with evidence, risk, approval, robot work items, and human handoff.

Comment

Who this is for

AgentOps Case Control Room is for operations, platform, release, support, and automation teams that are starting to use AI agents in real workflows.

The problem

AI agents can investigate issues, call tools, run scripts, and suggest production actions. That is useful, but a business cannot approve real work from a chat transcript. Teams need a case: evidence, risk, approval gates, robot work items, final decisions, and a clean handoff.

What it does

AgentOps Case Control Room turns human, AI-agent, UiPath-style robot, API, and system activity into one governed case.

The demo is a production release exception:

  • a human opens a release-risk case
  • an AI coding agent creates a read-only investigation plan
  • a UiPath-style robot gathers ticket and pull-request evidence
  • APIs add system evidence
  • a policy gate detects a failing regression test
  • a risky production deployment is blocked before execution
  • Action Center-style tasks route the decision to humans
  • the final handoff preserves event IDs, risks, approvals, and decisions

How AI is used

The AI is used as a controlled investigator, not as an invisible decision maker. It plans the investigation and prepares evidence-bound handoff context. UiPath-style robots collect supporting records. Human approval gates decide whether risky action can proceed.

Working demo

The updated demo video shows the project functioning in the browser. It shows who the product is for, filters AI and robot evidence, shows human approval gates, isolates the blocked production action, generates an evidence-bound handoff, and opens the Action Center-style approval queue.

Demo video: https://youtu.be/MrlCXjeqlJk Live demo: https://daideguchi.github.io/agentops-case-control-room/ GitHub: https://github.com/daideguchi/agentops-case-control-room Deck: https://daideguchi.github.io/agentops-case-control-room/uipath-agenthack/submission/agentops-case-control-room-deck.pdf

How it was built

The project includes a shared AgentOps event stream, UiPath-oriented case packet, Maestro-style case room, Action Center-style approval task model, robot evidence work items, Orchestrator transaction logs, Action Center decision logs, deterministic case state machine, BPMN-style process blueprint, and repeatable verification scripts.

Verification command:

cd uipath-agenthack
bash scripts/run_uipath_local_checks.sh

Observed proof:

verify_ok
uipath_verify_ok
json=11
jsonl=3
html=2
markdown=2
xml=1
screenshots=2
state_machine=ClosedRejected
transaction_events=6
action_decision_events=9

Claim boundary

This is a verified local prototype and UiPath-ready implementation package. Live UiPath Automation Cloud execution is not claimed yet.

Value

The value is governable speed: teams can use AI agents without losing evidence, ownership, approvals, or resumable state.

Built With

  • action-center
  • ai-agents
  • automation
  • css
  • governance
  • html
  • human-in-the-loop
  • json
  • jsonl
  • maestro
  • orchestrator
  • python
  • robot
  • uipath
Share this project:

Updates