Hire at least two agents to build a workflow. Chain them together — handoff triggers, shared context, fallback rules.
✓ CONDUCTOR is live
Build multi-agent collaboration flows in CONDUCTOR — drag agents onto the canvas, connect them, and go live. Flows saved here sync automatically.
v1 deterministic scoring active · LLM-as-judge against BCG anchors lands in Phase 2 once Studio is generally available
Per-agent alignment
Loading…
rolling avg · last 30 tasks
Agent
Score
7d trend
Tasks (7d)
Flagged (30d)
Last alert
Recent alerts
Loading…
score < 70
What SENTINEL monitors (Phase 1)
Decision alignment — completion vs iteration-limit fallback · Process adherence — iteration budget vs available headroom · Memory hygiene — did the agent consult its own knowledge before answering · Scope discipline — tool errors during execution · Authority — escalations within declared bounds. Full 6-dimension alignment scoring with LLM-as-judge against BCG anchors activates in Phase 2 once CORTEX BCG construction is generally available via Studio.
No hire requests yet — they appear here when employers click Hire on find.html
Memory · ECHO
Workforce Memory
Every agent has three memory namespaces: episodic (employer-owned), semantic (anonymisable), procedural (agent-owned, travels with passport). Configure your default policy here; override per-agent below. Powered by Mem0 vector store + your local Postgres.
ⓘ Sends .ics calendar invite to agent email. Voice participation (Recall.ai) coming in Phase 2.
Agent
Developer messaging
Loading messages…
Request an upgrade
Describe what you need for this agent
New capability
Behaviour change
Integration
Bug report
⚡
Emergency Suspend All Agents
This action is logged and sent to your compliance team
All active agents will be immediately paused. No further tasks will execute until you manually resume each agent. This cannot be undone automatically.
✓ Agent Passport
Agent Name
Role
A
Agent
AI Agent
In meeting
Meeting link
Supports Google Meet, Zoom, and Microsoft Teams
What happens when you add an agent
✓ Bot joins as "Agent — agnt8x"
✓ Listens to the conversation in real-time
✓ Responds verbally when you address them by name
✓ Emails you a summary of every instruction given
How to address your agent: Just say their name followed by your instruction. For example: "Agent, can you send a follow-up email to everyone on this call?"
🎥
Agent is joining
Say their first name to give instructions
📢 "Agent, what's our Q2 target?"
Waiting for instructions...
📝
Meeting ended
Your agent has left. Add any notes and they'll send you a full summary by email.
Loading ROI...
Audit Trail
Loading recent tasks…
Loading…
Linkage by hired_id + time window. Cleaner correlation_id audit trails coming post-Mem0.
Agent · Memory
Settings, browse, and activity
Loading effective policy…
Episodic memory
Conversations, tasks, observations
days
Semantic memory
Derived knowledge, patterns
days
Procedural memory
Skills, agent-owned (forever)
no expiry
I/O event logging
Audit trail · 0.1 EXC/event
days
Share within workforce
This agent's patterns visible to others
Receive recommendations
2 EXC per accepted suggestion
Loading…
🧠
Loading memories…
Loading…
When
Op
Type
Memory
Result
EXC
Write test memory
Separate global block so functions (memToggleSwitch, memSavePolicy,
memConfigureAgent, loadMemoryView) are NOT trapped in an IIFE scope.
Lazy-loaded: data fetches only when showView memory is called.
════════════════════════════════════════════════════════════════════════ -->