SYSTEM_CONSOLE v2.4.0

Agentic systems governance

How governance changes when agents chain tool calls, delegate to sub-agents, and act with partial autonomy.

LAST_UPDATED: 2025-05

While single-turn RAG and agentic systems share the same retrieval and policy fundamentals, agentic systems introduce compounding risks. Agents may chain tool calls across multiple steps, delegate to sub-agents, or accumulate privileges that no single step would have granted alone. Standard access control is necessary, but often insufficient.

Key Takeaways

  • • Policy gates must exist at every agent hop, not just at the entry point.
  • • Budget and iteration limits are safety controls, not just cost controls.
  • • Human approval gates are required for irreversible or high-impact actions.

Governed agent chain

Policy gates and budget controls across an orchestrator and sub-agent topology.

Governed Agent Chain

New risks introduced by agents

Privilege escalation across hops

Each agent step runs with its own identity. If sub-agents are not independently scoped, an orchestrator can grant access that no individual step would be allowed.

Prompt injection via tool outputs

A tool response containing adversarial text can redirect the agent's next step. Tool outputs must be sanitised before being re-injected into context.

Unbounded loops

Without hard iteration and cost limits, an agent can loop indefinitely or trigger cascading tool calls that consume unbounded resources.

Non-deterministic execution paths

The path an agent takes to complete a goal may vary across runs. Governance must capture the trace, not just the final output.

Per-hop policy model

Every agent handoff requires a new access decision. The orchestrator's permissions do not automatically flow to sub-agents, so each hop must be evaluated independently.

Control Applied where Why it matters
Identity scoping Each sub-agent Prevents privilege inheritance across the chain
Tool allow-list Per agent role Sub-agents only access tools they need for their task
Context boundary Between agents Prevents classified data leaking into untrusted downstream agents
Output validation Before re-injection Blocks adversarial content in tool results from redirecting the plan
Trace ID propagation Entire chain Links every step to the originating user request for auditability

Budget and iteration controls

Treat limits as safety rails, not just cost controls. A runaway agent is an operational incident.

Max iterations
Hard cap on agent loop steps per task
Token budget
Total tokens per task, across all hops
Tool call limit
Max external calls per agent execution
Wall-clock timeout
Hard deadline regardless of progress

Human approval gates

Not every action should be fully automated. Define which action types require a human confirmation before the agent proceeds.

Requires human approval

  • Writes or mutations to production systems
  • Actions on behalf of another user
  • Sending external communications
  • Actions above a defined cost threshold
  • Any irreversible action

Can proceed autonomously

  • Read-only retrieval within scoped sources
  • Summarisation and reformatting
  • Idempotent, low-impact tool calls
  • Draft generation (not delivery)
Best practice
Maintain a registry of action types with their approval requirements. This is an operational policy decision, not a model decision.

GCP mapping

Illustrative. Each layer maps to equivalent services on AWS, Azure, or any cloud.

Orchestration
Vertex AI Agent Builder / Cloud Run
Identity scoping
Workload Identity / per-agent SA
Budget control
Cloud Billing Budgets + Pub/Sub alerts
Trace store
Cloud Trace + BigQuery audit tables

Failure modes

  • ! Orchestrator's permissions are inherited by all sub-agents, bypassing per-hop scoping.
  • ! Tool output is re-injected without sanitisation, enabling prompt injection mid-chain.
  • ! No iteration limit allows runaway loops that exhaust budget or trigger rate limits.
  • ! Human approval gates are skipped under 'efficiency' pressure, enabling unsafe mutations.
  • ! Agent traces are not linked to user identity, making post-incident investigation impossible.

Checklist

  • Each agent hop has its own scoped identity and tool allow-list.
  • Tool outputs are validated before re-injection into agent context.
  • Hard limits exist for iterations, tokens, tool calls, and wall time.
  • A human approval gate registry defines which actions require sign-off.
  • Every agent execution trace is linked to the originating user and request.