Skip to content
Cadre
Technology agents
Enterprise OperationsTechnologyAI / Agent Platform (AgentOps)

Agent Registry & Protocol Agent

Maintains the A2A, MCP and AP2 service registry and versioned capability records.

Registers agent capabilities and tools, brokers approved service discovery, and maintains versioned prompts and tool schemas. It resolves a request to an eligible, version-pinned role and validates the mandate required for agent-initiated payments.

Authority

Act within policy

Team role

Routes work to specialists

Handoffs

Named collaborators

The role

What it owns and where its authority ends

Desk

AI / Agent Platform (AgentOps)

Desk workflow

Register, then version, then regression evals, then champion/challenger release, then live tracing and guardrails, then drift detection, offline consolidation and re-evaluation. The agentic control plane governs the fleet; the board sets the mandate and holds the kill-switch.

Collaboration

Works within a defined desk workflow

Decision boundary

Acts only within an explicit policy, permission and escalation boundary.

Systems and capabilities involved

  • A2A capability-card registry

  • MCP server discovery

  • Prompt / tool-schema version store

  • AP2 mandate validation

    agent-payment authorization

Handoffs

What this role gives and receives

Capabilities offered

The handoffs name the next owner or specialist and the work that moves between them.

Receives from

External handoff

Treasury for agent-payment mandate limits

Context

What the role needs to do the work

Current work
The registration/version change being validated.
Prior interactions
Not specified for this role.
Policies and reference
MCP/A2A/AP2 specs, capability cards, the prompt/tool version graph.
Working method
Not specified for this role.

Checks and boundaries

What must be tested or reviewed

  1. 01Schema + capability-card validation on every registration; malformed entries quarantined.
  2. 02Version pins are immutable and reversible; no agent resolves to an un-evaled version.
  3. 03AP2 mandates checked against scope/limit policy before any agent-payment edge is allowed.

Human authority

Acts only within an explicit policy, permission and escalation boundary.

Keep exploring

More agents in AI / Agent Platform (AgentOps)

Eval Harness AgentRuns regression, independent-review and adversarial checks before an agent release.Fleet Observability AgentMonitors production agents for drift, failures, cost and control signals.Guardrails & Kill-Switch AgentEnforces policy controls and coordinates scoped intervention.