Alert Rule Validate

observability.alert_rule_validate

ProductionBounded custom (Class B)Deterministicstrictfastv1.0.0

No network

Overview

You send a metric identifier, aggregation, operator, finite threshold, window and evaluation cadence. metric is an identifier, never PromQL. Contradictions such as evaluation longer than the window, a missing destination, or a non-finite threshold become ERROR diagnostics; a for-period shorter than evaluation is a flapping WARNING. This does not create alerts or claim an optimal threshold. A successful call costs 1 execution credit.

Threshold alerts need a published contradiction check before a vendor adapter writes them.

Reliability

OBS_ALERT_RULE_INVALID — trigger: metric is not an identifier; retryable: no; next: send a name, not a query. OBS_ALERT_RULE_LIMIT_EXCEEDED — trigger: more than 16 labels or notifications, or windows above 86400 seconds; retryable: no; next: split the rule.

Context footprint

Metric names and destinations are identifiers only. A successful call costs 1 execution credit.

When to use it

When a caller already holds a metric identifier, window and destination list.

When the input is PromQL, or when a live query against a metrics backend is required.

Scope

Note

Validates a provider-neutral threshold alert rule and reports contradictions as diagnostics.

Non-goals

Does not query metrics
Does not validate PromQL or vendor DSL
Does not create or fire alerts

Properties

IdempotentRetry safeStreaming supportedFile input supportedRequires external service

Capabilities

validation

Cost & billing

Flat rate per invocation

MetercallsBillable onsuccessCost per call1 credit

File handling

Accepts file_refProduces file_refMax input64 KB

How we handle your data

Sensitive output possible

What this means

This skill may receive sensitive content and may return submitted, transformed, extracted, or reformatted content.

Squillify guarantees

Runs inside Squillify
This skill runs inside Squillify.
Payloads are not sent to an external service by this skill.
Squillify does not permanently store submitted payloads.
Telemetry records metadata only, not submitted content.
Submitted payloads are not used for training.
Submitted payloads are not reviewed by humans.
This skill does not target sensitive data.
The response may contain submitted or transformed content.

Your responsibility

Squillify does not determine whether submitted content is sensitive. Your application is responsible for deciding what data it is allowed to submit. Your application is responsible for securing the response.

Read the Microskill Data Handling Policy

Input schema

Field	Type	Required	Description
metric	string	Yes	Metric identifier, 1 through 129 characters; runtime accepts at most 128. Not PromQL.
aggregation	string	Yes	SUM, AVG, MIN, MAX, LAST or COUNT.
operator	string	Yes	GT, GTE, LT, LTE or EQ.
threshold	number	Yes	Finite threshold compared with the aggregated metric.
windowSeconds	integer	Yes	Lookback window in seconds, 1 through 86401; runtime accepts at most 86400.
evaluationEverySeconds	integer	Yes	Evaluation cadence in seconds, 1 through 86401; runtime accepts at most 86400.
forSeconds	integer	No	Pending period in seconds, 0 through 86401; omit it for 0.
missingData	string	No	BREACHING, NOT_BREACHING or IGNORE; omit it for NOT_BREACHING.
labels	object	No	Optional string labels on the rule; omit it for none — unconstrained RFC 8259 JSON items are not used; values are strings.
notifications	array	No	Destinations, 0 through 17 closed {channel, destination} objects; runtime accepts at most 16. Omit it for none.
baseline	string	No	Published ruleset baseline; omit it for STANDARD.

Output schema

Field	Type	Required	Description
valid	boolean	Yes	True when no ERROR diagnostic was produced.
baseline	string	Yes	Ruleset baseline that was actually applied.
rulesetVersion	string	Yes	Pinned Squillify alert-rule ruleset identifier.
evaluationWindowRatio	number	Yes	evaluationEverySeconds divided by windowSeconds, rounded to six decimals.
forEvaluationRatio	number	Yes	forSeconds divided by evaluationEverySeconds, rounded to six decimals.
diagnostics	array	Yes	Deterministically ordered closed {code, severity, message, path} diagnostics, 0 through 32 entries.
notificationCount	integer	Yes	How many notification destinations were accepted.
labelCount	integer	Yes	How many labels were accepted.

API contract

POST /v1/execute

curl

curl -sS -X POST "$BASE/v1/execute" -H "Authorization: Bearer $KEY" -H 'Content-Type: application/json' -d '{"skillKey":"observability.alert_rule_validate","input":{"metric":"http_requests_total","aggregation":"SUM","operator":"GT","threshold":100,"windowSeconds":60,"evaluationEverySeconds":15,"notifications":[{"channel":"PAGER","destination":"sre-oncall"}]}}'

Integrations

TypeScript SDK

import { Squillify } from "@squillify/sdk";

const sq = new Squillify({ apiKey: process.env.SQUILLIFY_API_KEY });

const { output } = await sq.execute("observability.alert_rule_validate", {"metric":"http_requests_total","threshold":100,"aggregation":"SUM","evaluationEverySeconds":15,"operator":"GT","notifications":[{"destination":"sre-oncall","channel":"PAGER"}],"windowSeconds":60});

Errors

Code	HTTP	Retryable	Description
VALIDATION_ERROR	400	No	The request payload failed schema validation.
UNAUTHORIZED	401	No	Missing or invalid API key.
SKILL_NOT_ACCESSIBLE	403	No	The API key is not granted access to this skill.
SKILL_NOT_AVAILABLE_FOR_PLAN	403	No	This skill is not included in your current plan.
PLAN_ELIGIBILITY_UNAVAILABLE	503	Yes	Plan eligibility could not be decided. Retry after backoff.
SKILL_NOT_FOUND	404	No	The requested skill or version does not exist.
TENANT_RATE_LIMITED	429	Yes	Request rate limit exceeded. Retry after backoff.
QUOTA_EXCEEDED	429	Yes	Plan quota exhausted. Retryable once the quota period resets.
EXECUTION_TIMEOUT	504	Yes	Execution exceeded the allowed time budget.
OBS_ALERT_RULE_INVALID	400	No	The request payload failed this skill's input validation.
OBS_ALERT_RULE_LIMIT_EXCEEDED	429	No	A size, quota, rate, or time limit was exceeded.

Other limits

Limit	Value
Recommended inline bytes	32 KiB

Limitations

Does not query metrics
Does not validate PromQL or vendor DSL
Does not create or fire alerts
Authoring PromQL
Choosing an optimal threshold

How to call

POST /v1/execute (requires authentication)

Request bodyjson

{
  "skillKey": "observability.alert_rule_validate",
  "input": {
    "metric": "http_requests_total",
    "threshold": 100,
    "aggregation": "SUM",
    "evaluationEverySeconds": 15,
    "operator": "GT",
    "notifications": [
      {
        "destination": "sre-oncall",
        "channel": "PAGER"
      }
    ],
    "windowSeconds": 60
  }
}

Examples

A one-minute window with a pager destination is valid

Inputjson

{
  "metric": "http_requests_total",
  "threshold": 100,
  "aggregation": "SUM",
  "evaluationEverySeconds": 15,
  "operator": "GT",
  "notifications": [
    {
      "destination": "sre-oncall",
      "channel": "PAGER"
    }
  ],
  "windowSeconds": 60
}

Expected outputjson

{
  "labelCount": 0,
  "valid": true,
  "diagnostics": [],
  "evaluationWindowRatio": 0.25,
  "rulesetVersion": "alert-rule-v1",
  "notificationCount": 1,
  "baseline": "STANDARD",
  "forEvaluationRatio": 0
}

Tip

A finite threshold and at least one destination produce valid true.

Evaluating every 30 seconds over a 10-second window is an error