2026-05-12 Kade Cowper

Policy as Code for AI Agent Enforcement: What It Means and How It Works

Policy as code is arriving in AI agent governance — Kyndryl, Microsoft, and Altimetrik are all publishing on it in 2026. Most of what they describe is declaration: expressing agent policy as structured configuration rather than prose. The part most implementations leave out is enforcement: a runtime gate that evaluates the policy against every action before execution and produces a signed receipt as proof.

The Problem With Prompt-Based Policy

Most AI agent "policies" today live in the system prompt. The prompt tells the agent what it's allowed to do, what steps to follow, what data it may access. This is policy-as-natural-language: readable by humans, interpretable by the model, unenforceable by anything.

A system prompt is a probabilistic instruction. The model reads it and approximates compliance on each generation. There is no mechanism that prevents a non-compliant action from executing — if the model generates a bad output, the action runs. If a prompt injection overwrites the instruction, the agent follows the injection. If the model simply hallucinates a step that isn't in the policy, nothing stops it.

Policy expression Machine-readable Versioned Enforced before execution Auditable evidence
System prompt / guidelines No — natural language No — lives in model context No — probabilistic No — model output only
Post-hoc evaluation (LLM judge) Partial — structured prompt Partial — prompt versioning No — runs after execution Partial — evaluation records
Policy as code — enforcement gate Yes — structured contract Yes — versioned with deployment Yes — gate fires pre-execution Yes — signed receipt per decision

What Agent Policy as Code Actually Is

Policy as code in the DevOps sense (Open Policy Agent, HashiCorp Sentinel) evaluates infrastructure configuration at deploy time — checking whether a proposed change conforms to policy before it is applied. That is useful, but it is a one-shot check at a single point in a deployment pipeline.

Agent policy as code is a continuous runtime check. The policy is evaluated on every action, during a live session, in the critical window between the agent's decision to act and the action's execution. The enforcement point is not "before deploy" but "before each tool call."

Three components are required:

1 — Step Order Contract
  • Declares the permitted sequence of steps
  • Machine-readable string or config
  • Versioned with the enforcement worker
  • Violations → SEQUENCE_VIOLATION
2 — Action Type Map
  • Declares which action categories each function permits
  • Versioned policy map (not prompt instructions)
  • Violations → ACTION_NOT_ALLOWED
  • Unknown functions → NO_POLICY_MATCH
3 — Enforcement Gate
  • Independent of the agent — cannot be overridden
  • Evaluates policy before execution
  • Returns ALLOW or DENY
  • Writes signed receipt at decision time

The policy is not "what the agent believes it should do." The policy is what the gate evaluates against. The agent's understanding is irrelevant — the gate's decision is authoritative.

AgenticRail's Sequence Contract

AgenticRail implements policy as code as a two-part contract: the step order and the function/action_type map.

The step order is declared as an ordered list. For an 8-step MSMD spine:

Step order policy — versioned environment variable
SLP8_STEP_ORDER_MSMD="intake,disruption,instability,state_read,internal_driver,execution,boundary,settle"

This string is the policy. It is stored as an environment variable on the enforcement worker, deployed as part of the worker version, and auditable via the deployment history. Every receipt written by the gate references the key_id of the signing key in effect at decision time — tying the receipt to the policy version that produced it.

The function/action_type map declares which action categories each function permits:

Policy map — function → permitted action types
// intake permits state checks — not writes
"intake": ["CHECK_STATE", "READ_INPUT"]

// execution permits writes and external calls
"execution": ["WRITE_DB", "CALL_EXTERNAL_API", "EMIT_EVENT"]

// settle permits only finalisation
"settle": ["SEAL_SEQUENCE", "WRITE_SUMMARY"]

An agent attempting to call WRITE_DB during intake — regardless of what its prompt or reasoning said was appropriate — receives DENY: ACTION_NOT_ALLOWED before the write executes.

The Enforcement Gate: Where Policy Becomes Code

The gate is what separates policy-as-code from policy-as-documentation. Without a gate, you have a policy document. With a gate, you have enforcement.

The gate sits between the agent and every downstream action. The agent does not call tools directly — it calls the gate, which evaluates the declared policy and decides whether the action may proceed. The agent cannot route around the gate; it cannot modify the policy at runtime; it cannot bypass the nonce check or the step order constraint.

Key property

The agent is the regulated system. The gate is the regulator. A system cannot regulate itself — the independence of the gate is what makes the policy enforceable rather than advisory.

Every gate decision produces a signed receipt, written before the action executes:

ALLOW receipt — policy check passed, action authorised
{
  "sequence_id": "loan-approval-20260512-001",
  "step": "execution",
  "function": "execution",
  "action_type": "WRITE_DB",
  "decision": "ALLOW",
  "policy_version": "k1_2026-02-22_01",
  "ts_ms": 1715508000000,
  "hmac_sig": "sha256=9e2f4b..."
}

Denial Codes: What Policy Violations Look Like

When the policy is violated, the gate returns a specific reason code. Each code maps to a distinct policy rule:

SEQUENCE_VIOLATION
Step submitted out of declared order. Step order policy enforced.
ACTION_NOT_ALLOWED
Action type not in the permitted set for this function.
NO_POLICY_MATCH
Function name not recognised in the policy map.
REPLAY_NONCE
Nonce already used in this session. Replay protection policy enforced.
SEALED_SEQUENCE
Session reached final step and was sealed. No further actions permitted.
STALE_TIMESTAMP
Request timestamp outside the 5-minute freshness window.

Each denial produces a signed receipt — immutable, pre-execution evidence that the policy ran and what it rejected. These receipts are what compliance auditors, EU AI Act Article 9 documentation, and ISO 42001 evidence packages require.

Version Control and Auditability

Because the policy is expressed as code — not embedded in a prompt — it has all the properties of code:

6 Specific DENY reason codes
Pre-exec Receipt written before action runs
Versioned Policy stamped into every receipt

Compliance Mapping

Framework Requirement How policy as code satisfies it
EU AI Act Article 9 Risk management system — identify, analyse, mitigate risks Policy contract is the declared risk boundary; gate receipts prove it was enforced
EU AI Act Article 12 Automatic logging enabling reconstruction of events Pre-execution receipts are the reconstruction anchors — not post-hoc observations
ISO/IEC 42001 A.6.1.6 Controls for AI system operation — documented and evidenced Policy document + receipt chain = documented control + evidence it ran
NIST AI RMF Govern 1.2 Accountability mechanisms for AI system decisions Every gate decision attributed to a policy version, signed, and stored immutably
OWASP ASI01 (Goal Hijack) Prevent agent from being redirected to unauthorised objectives Step order and action type policy enforced externally — prompt injection cannot override

The Part Most Implementations Skip

The industry conversation about policy as code for AI agents tends to stop at declaration. Expressing agent governance as structured configuration rather than prose is valuable — it is readable by tools, comparable across versions, and unambiguous in intent.

But declaration without enforcement is documentation. A policy document does not prevent a non-compliant action from executing. It describes what should happen; it does not guarantee what does happen.

The enforcement gate is what makes the code operative. Without it, "policy as code" is a better way of writing the same unenforceable guidelines that used to live in the system prompt.

See policy as code in action

Run a sequence through the AgenticRail gate — submit a step out of order and inspect the SEQUENCE_VIOLATION receipt.

Open Demo Read Docs
← All posts