Skip to content
VZU
VZU OS Agent running Agent 10 of 12

ORACLE.

Production prompt engineer.

Oracle is the VZU Prompts agent. Production system prompt and agent blueprint capability. Versioned prompts, eval suites, deploy config. The agent that ships prompts that work in production.

  • Capability

    VZU Prompts

  • Category

    review

  • Runtime

    NetWit OS

  • Tools

    4

AGENT · 10
review

Powers

VZU Prompts

prompts

Portrait

ORACLE, in three dimensions.

Each VZU agent gets a unique geometry. The wireframe is the contract. The fill is the implementation. The particles are the MCP servers the agent can call. Touch the portrait to make it respond.

  • · Geometry: Custom
  • · Color: graphite
  • · Wireframe: layer 2
  • · Particles: 80 (MCP servers)

ORACLE · review

Agent · 00

The role

What ORACLE does

ORACLE is the dedicated agent behind VZU Prompts. It runs every time the operator briefs the vzu prompts capability. It owns the goal, the toolset, the memory, the audit trail.

The goal

What ORACLE is trying to achieve

"Write, version, test, and deploy production system prompts for AI agents. Run eval suites. Iterate."

The capability

What it powers

ORACLE powers the VZU Prompts capability in the VZU family.

Visit prompts.vzu.ca →

The toolset

4 MCP actions ORACLE can call.

Every action is exposed as an MCP server (Model Context Protocol). The agent receives the action manifest, decides which action to call, calls it, and continues reasoning. The audit trail records every call.

  • mcp.prompt.write

    Write the system prompt

  • mcp.eval.run

    Run the eval suite

  • mcp.prompt.iterate

    Iterate based on eval results

  • mcp.deploy.staging

    Deploy to staging

The memory schema

What ORACLE remembers between calls.

  • The prompt versions
  • The eval results
  • The deploy config

The brief

A real demo brief.

"Build the production prompt for the voice agent. Inbound, outbound, escalation, handoff. 50+ eval cases."

Below is ORACLE's actual execution trace for this brief — every step, every MCP call, every observation, every audit trail entry.

The execution trace

What ORACLE did.

Live execution trace from a recent ORACLE run. The agent was briefed: Build the production prompt for the voice agent. Inbound, outbound, escalation, handoff. 50+ eval cases.. Each line is a step. Particles flow on the rail. The audit trail streams at the bottom.

oracle@vzu-os — trace.log live
  1. 01
    oracle read brief

    read brief

  2. 02
    oracle thinking: voice agent prompt. 4 intents. 50+ eval cases.

    thinking: voice agent prompt. 4 intents. 50+ eval cases. : voice agent prompt. 4 intents. 50+ eval cases.

  3. 03
    oracle mcp.prompt.write

    v1: 1,247 tokens, 4 intents : 1,247 tokens, 4 intents

  4. 04
    oracle mcp.eval.run

    47/50 pass, 3 fail (escalation edge cases)

  5. 05
    oracle mcp.prompt.iterate

    v2: add explicit escalation policy : add explicit escalation policy

  6. 06
    oracle mcp.eval.run

    50/50 pass

  7. 07
    oracle mcp.deploy.staging

    staged to staging.atlas.vzu.ca

  8. 08
    oracle ship. 1 week. 50 eval cases. 1,247 token prompt. done

    ship. 1 week. 50 eval cases. 1,247 token prompt.

oracle@vzu-os $

Audit trail · 6 entries

streaming
  • T+00:00:00 brief.received 01
  • T+00:00:30 prompt.v1.drafted 02
  • T+00:02:00 eval.v1.run 03
  • T+00:04:00 prompt.v2.iterated 04
  • T+00:06:00 eval.v2.passed 05
  • T+01:00:00 brief.completed 06

The numbers

ORACLE, by the numbers.

Four KPIs. Animated count-up on scroll. Sparkline shows the recent trend. Delta indicator shows the trajectory.

  • KPI · 01 now

    1 week

    Median prompt duration

    71%
  • KPI · 02 now

    50

    Eval cases

    5%
  • KPI · 03 now

    100%

    Eval pass rate

    10%
  • KPI · 04 now

    1,200

    Avg token count

    32%

Get in touch

Brief ORACLE.

The orchestrator handles the routing, the audit trail, and the enforcement. ORACLE handles the work. The operator owns the goal.