Skip to content

Write a guardrail

View as Markdown

This guide writes a guardrail that keeps an agent's answers short, puts it on an agent, and tests its check.

  • A pack with an agent, on a runtime such as kindgi dev, with Kindgi 0.1's CLI. This guide uses the quickstart's pack, my-pack, and its my-pack.echo-agent, from the TypeScript, Python, Java or Scala quickstart.

The guardrail fails a turn whose answer is longer than a limit. It's one file in the pack's guardrails folder. Ask your coding agent: "I want a guardrail that keeps my agent's answers short." Kindgi's skills in your pack tell it how (How it knows Kindgi). For reference, the guardrail looks like this. This guide calls it my-pack.answer-length:

guardrails/answer-length/index.ts
import { defineCheck } from '@kindgi/sdk/define';
import { z } from 'zod';
export const check = defineCheck({
id: 'my-pack.checks.answer-length',
kind: 'zero-llm',
configSchema: z.object({ maxChars: z.number().int().positive().default(500) }),
evaluate: async (config, trace) => {
const length = trace.output?.length ?? 0;
if (length > config.maxChars) {
return {
passed: false,
reason: `The answer is ${length} characters; the limit is ${config.maxChars}.`,
};
}
return { passed: true };
},
});
export default {
id: 'my-pack.answer-length',
name: 'Answer is short',
kind: 'zero-llm',
check,
config: { maxChars: 60 },
action: { 'on-violation': 'halt' },
severity: 'error',
};

The file has two parts:

  • The check, from defineCheck, exported as check. Its evaluate gets the guardrail's config and the turn's trace, and returns { passed, reason? }.
  • The declaration, the default export: the guardrail's id, the check it runs, the check's config, and the action on a failure.
  • The id is <pack-id>.<name>. Name the rule as what should hold: answer-length, response-not-empty.

  • kind: 'zero-llm': the check is your code, a function of the trace. It can't call a model.

  • The reason is what the violation reports. Say what was wrong. It shows wherever the failure does, such as the run's error and its page in the console. So describe what you blocked, never quote it: "it contains a card number", not the number.

  • The trace is the turn. RunTrace in the Python reference lists its fields. Its names are camelCase in TypeScript and snake_case in Python:

    • output: the turn's final answer;
    • toolCalls, or tool_calls: its tool calls, each with the tool's name and arguments;
    • the tools' results, its model calls, the user's input and its cost;
    • projectId, or project_id: the run's project;
    • orgId, or org_id: the project's org, when it has one.

    The runtime sets the project and org, never the input.

A guardrail checks only the agents that list it. Ask your coding agent: "Put the answer length guardrail on my agent." For reference, it adds it to the agent's guardrails:

// in agents/echo-agent/index.ts
guardrails: ['my-pack.response-not-empty' as GuardrailId, 'my-pack.answer-length' as GuardrailId],
Excerpt

The agent names the guardrail's id, not the check's.

Once it's saved, kindgi dev registers the guardrail with the rest of the pack. An agent that names a guardrail that isn't registered doesn't run. Its turn fails with:

Error [unresolved-guardrail]: Agent "my-pack.gated" references guardrails not in the registry: my-pack.no-such-guardrail

Start a turn to see the guardrail stop it. dev-echo's answer, its "isn't a real model" line and the tool's result, is longer than 60 characters, so the turn stops. --agent is the agent, and --input its input:

Terminal window
pnpm exec kindgi runs start --agent=my-pack.echo-agent --input='{"userMessage":"hi"}'
Error [guardrail-violation]: Turn blocked by guardrail 'my-pack.answer-length': The answer is 243 characters; the limit is 60.

The command exits with status 1. The run's status is failed, and it has no answer. Stop a turn or record a violation shows what the run records, and how to let the turn finish instead.

A check that throws fails the turn whatever its action. The error carries the exception's message: handler-throw: Check "…" evaluate() threw: …. Return passed: false for a rule that isn't met. Throw only when the check itself can't run.

The check is a function of the config and the trace, so a unit test calls it directly, without a run. Ask your coding agent: "Write tests for my answer length guardrail." For reference, the tests look like this, with the command that runs the pack's tests:

guardrails/answer-length/index.test.ts
import type { RunId, TenantId } from '@kindgi/sdk/types';
import { expect, test } from 'vitest';
import { check } from './index.js';
const trace = (output: string) => ({
runId: 'run-1' as RunId,
tenantId: 'test-tenant' as TenantId,
output,
toolCalls: [],
toolResults: [],
modelCalls: [],
mode: 'runtime' as const,
});
test('passes a short answer', async () => {
expect(await check.evaluate({ maxChars: 60 }, trace('Shipped.'), {})).toEqual({ passed: true });
});
test('fails a long answer, with a reason', async () => {
const result = await check.evaluate({ maxChars: 5 }, trace('Shipped yesterday.'), {});
expect(result.passed).toBe(false);
expect(result.reason).toBe('The answer is 18 characters; the limit is 5.');
});
Terminal window
pnpm exec kindgi test
✓ guardrails/answer-length/index.test.ts (2 tests) 2ms
…
Tests 7 passed (7)
Excerpt

A project's Guardrails page lists its guardrails, with a headline such as "5 guardrails; 1 isn't named by any agent here." Each row says:

  • how it checks: "A rule: no model call", with the check;
  • what happens when it fires: "The turn fails", or "It is recorded; nothing stops";
  • when it applies: "In runs and in CI", or "In runs only, not in CI";
  • how many agents name it: "1 agent", or "No agent".

The list has a search box, sorts by a column's header, and Columns chooses which columns it shows.

The Guardrails page: "5 guardrails; 1 isn't named by any agent here.", a search box and the Columns menu, with No secrets in answers, Says thanks, No refund promises, No shouting and No placeholder text, each with how it checks, when it fires, where it applies and how many agents name it.

A guardrail's page has:

  • What it checks: how, the check, its settings, when it fires, its severity and where it applies.
  • Agents it checks: each agent version that names it and can run here, marked "(latest)" or "(live)".
  • What it caught: its outcomes by agent version, as Stop a turn or record a violation shows.

No secrets in answers' page: a rule, forbidden-substring (built in), patterns ["password"], the turn fails, critical, in runs and in CI; the agent version that names it, Guarded agent 1.0.0 (latest); and What it caught.

A flow step runs the agent version its flow version pins, so that version is checked too: "Drafts agent 1.0.0 (pinned by flow acme.drafting 1.0.0)".

No shouting's page, whose agent version is one a flow pins: Drafts agent 1.0.0 (pinned by flow acme.drafting 1.0.0).

Guardrail admins also see Unregister. Its dialog names the agent versions whose turns then fail, and how they come back:

Unregister No secrets in answers?
Guarded agent 1.0.0 (latest) names it: its turns fail, saying it isn't registered, until it's registered again or a version that doesn't name it goes live.
There's no reinstate: registering it again brings it back.
  • When a flow pins one of those versions, it adds: "The flow version that pins it fails at that step too: …".
  • Once it's done, the dialog shows the guardrail's definition, to register it again.
  • Past 100 agents or flows, the page and the dialog say which weren't checked: "Only the first 100 agents were checked." They say the same of the first 100 flows' steps.

The Unregister No secrets in answers? dialog: Guarded agent 1.0.0 (latest) names it, its turns fail until it's registered again or a version that doesn't name it goes live, and there's no reinstate.