> **Building with AI coding agents?** Install the authstack plugin with one command. This equips your agent with accurate Scalekit implementation patterns.
>
> **Recommended** (the `-y` flags skip prompts, so the command runs without an interactive terminal):
> ```bash
> npx -y @scalekit-inc/cli setup -y
> ```
>
> Global:
> ```bash
> npm install -g @scalekit-inc/cli
> scalekit setup -y
> ```
>
> Supports Claude Code, Cursor, GitHub Copilot, Codex + skills for other Agent Skills-compatible agents in `~/.agents/skills`.
> Skills: integrate-agentkit, implement-saaskit, add-mcp-oauth, implement-sso, implement-scim.
> [Full setup guide](https://docs.scalekit.com/dev-kit/build-with-ai/)

---

# Connect AI agents to New Relic

The New Relic connector lets your AI agent act in each user's New Relic account. Each user connects their own New Relic API key once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 29 tools.

**Authentication:** API Key
**Categories:** Monitoring, Developer Tools
**Tools:** 29: 13 read, 12 write, 4 destructive
**Users sign in with:** API key
**Built by:** Scalekit connector
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=NEWRELIC)

## Setup

1. ### Install the SDK

   **Node.js**

   ```bash
   npm install @scalekit-sdk/node dotenv
   ```

   **Python**

   ```bash
   pip install scalekit-sdk-python python-dotenv
   ```

2. ### Set your credentials

   Add your Scalekit credentials to your `.env` file. Find values in **[app.scalekit.com](https://app.scalekit.com)** > **Developers** > **API Credentials**.

   ```sh title=".env"
   SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
   SCALEKIT_CLIENT_ID=<your-client-id>
   SCALEKIT_CLIENT_SECRET=<your-client-secret>
   ```

3. ### Create the New Relic connection

   In **AgentKit > Connections**, create a New Relic connection. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

4. ### Authorize a user and make your first call

   **Node.js** (`quickstart.mts`)

   ```typescript
   import { ScalekitClient } from '@scalekit-sdk/node'
   import 'dotenv/config'
   import { createInterface } from 'node:readline/promises'

   const scalekit = new ScalekitClient(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET,
   )
   const actions = scalekit.actions

   const connector = 'newrelic'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize New Relic:', link)
   const rl = createInterface({ input: process.stdin, output: process.stdout })
   await rl.question('Press Enter after authorizing...')
   rl.close()

   // Make your first call
   const result = await actions.executeTool({
     connector,
     identifier,
     toolName: 'newrelic_accounts_list',
     toolInput: {},
   })
   console.log(result)
   ```

   ```bash
   npx tsx quickstart.mts
   ```

   **Python** (`quickstart.py`)

   ```python
   import os
   from scalekit import ScalekitClient
   from dotenv import load_dotenv
   load_dotenv()

   scalekit_client = ScalekitClient(
       env_url=os.getenv("SCALEKIT_ENVIRONMENT_URL"),
       client_id=os.getenv("SCALEKIT_CLIENT_ID"),
       client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
   )
   actions = scalekit_client.actions

   connection_name = "newrelic"
   identifier = "user_123"

   # Generate an authorization link for the user
   link_response = actions.get_authorization_link(
       connection_name=connection_name,
       identifier=identifier,
   )
   print("Authorize New Relic:", link_response.link)
   input("Press Enter after authorizing...")

   # Make your first call
   result = actions.execute_tool(
       tool_input={},
       tool_name="newrelic_accounts_list",
       connection_name=connection_name,
       identifier=identifier,
   )
   print(result)
   ```

   ```bash
   python quickstart.py
   ```

   Each user opens the link once and enters their New Relic credentials there. See [Authorize a user](/agentkit/tools/authorize/) for the full flow and statuses.

## Tools

Pass the exact name to `execute_tool`, with an input like each tool's example. To ask for a tool that's missing, use the [request form](https://scalekitsupport.portal.usepylon.com/forms/request-a-connector-tool).

**Node.js**

```typescript
const result = await actions.executeTool({
  toolName: 'newrelic_accounts_list',
  toolInput: {},
  connector: 'newrelic',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="newrelic_accounts_list",
    tool_input={},
    connection_name="newrelic",
    identifier="user_123",
)
```

`result.data` is the app's response as JSON, and `result.execution_id` (`executionId` in Node.js) is the ID of the call.

### `newrelic_accounts_list`

List Accounts · Read-only

List the New Relic accounts the API key's user can access, plus that user's own profile.
Returns each account's numeric id and name, and the current user's id, name and email.
Use this first to find the account_id that the NRQL, alert, dashboard and synthetics tools need.
NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `newrelic_alert_issues_list`

List Alert Issues · Read-only

List alert issues (groups of related incidents) in a New Relic account, filtered by state, priority, entity or time window.
Returns a nextCursor for paging and each issue's issueId, title, description, priority, state, created/activated/closed/acknowledged times, acknowledgedBy, affected entity GUIDs and names, conditionName, policyName, totalIncidents and deepLinkUrl.
Use this to see what is alerting now or what fired in a time range. Use newrelic_alert_issue_acknowledge or newrelic_alert_issue_resolve to act on an issue.
Requires an account_id from newrelic_accounts_list. Without a time window New Relic returns the last 24 hours. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `cursor` (`string`, optional): Pagination cursor. Pass the nextCursor value from the previous response to get the next page; leave empty for the first page. Example: Mjox
- `end_time` (`integer`, optional): End of the time window, in Unix epoch milliseconds. Set it together with start_time. Example: 1767312000000
- `entity_guids` (`array`, optional): Return only issues that affect these entities. Get GUIDs from newrelic_entities_search. Example: ["MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg"]
- `issue_ids` (`array`, optional): Return only these issues. Example: ["f0846f5e-fc9d-4dc0-afdd-07cce7cc6fc4"]
- `priority` (`array`, optional): Return only issues with these priorities: LOW, MEDIUM, HIGH or CRITICAL. Example: ["CRITICAL", "HIGH"]
- `start_time` (`integer`, optional): Start of the time window, in Unix epoch milliseconds. Set it together with end_time; if either is missing New Relic uses the last 24 hours. Example: 1767225600000
- `states` (`array`, optional): Return only issues in these states: CREATED, ACTIVATED (open), DEACTIVATED or CLOSED. Example: ["ACTIVATED"]

Example input: `{"account_id":1234567}`

### `newrelic_alert_nrql_condition_get`

Get NRQL Alert Condition · Read-only

Get one NRQL alert condition by ID.
Returns its name, enabled flag, policyId, type, NRQL query, thresholds, description, runbookUrl, titleTemplate, signal settings (aggregation window, method, delay, fill option) and loss-of-signal expiration settings.
Use this to read a condition's full configuration before changing it. Use newrelic_alert_nrql_conditions_list to find condition IDs.
Requires a condition ID from newrelic_alert_nrql_conditions_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `condition_id` (`string`, required): ID of the alert condition. Get it from newrelic_alert_nrql_conditions_list. Example: 34567890

Example input: `{"account_id":1234567,"condition_id":"<condition_id>"}`

### `newrelic_alert_nrql_conditions_list`

List NRQL Alert Conditions · Read-only

List NRQL alert conditions in a New Relic account, optionally filtered by policy, name or query text.
Returns totalCount, a nextCursor for paging, and each condition's id, name, enabled flag, policyId, type (STATIC, BASELINE or OUTLIER), NRQL query, thresholds, description and runbookUrl.
Use this to find conditions or see what a policy alerts on. Use newrelic_alert_nrql_condition_get for one condition's full signal and expiration settings.
Requires an account_id from newrelic_accounts_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `cursor` (`string`, optional): Pagination cursor. Pass the nextCursor value from the previous response to get the next page; leave empty for the first page. Example: Mjox
- `name_like` (`string`, optional): Return only conditions whose name contains this text. Example: error rate
- `policy_id` (`string`, optional): Return only conditions in this alert policy. Get the ID from newrelic_alert_policies_list. Example: 4567890
- `query_like` (`string`, optional): Return only conditions whose NRQL query contains this text. Example: FROM Transaction

Example input: `{"account_id":1234567}`

### `newrelic_alert_policies_list`

List Alert Policies · Read-only

List alert policies in a New Relic account, optionally filtered by name.
Returns totalCount, a nextCursor for paging, and each policy's id, name, incidentPreference, accountId and entityGuid.
Use this to find a policy ID. Use newrelic_alert_policy_get for one policy and newrelic_alert_nrql_conditions_list for the conditions in a policy.
Requires an account_id from newrelic_accounts_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `cursor` (`string`, optional): Pagination cursor. Pass the nextCursor value from the previous response to get the next page; leave empty for the first page. Example: Mjox
- `name_like` (`string`, optional): Return only policies whose name contains this text (case-insensitive). Example: checkout

Example input: `{"account_id":1234567}`

### `newrelic_alert_policy_get`

Get Alert Policy · Read-only

Get one New Relic alert policy by ID.
Returns the policy's id, name, incidentPreference, accountId and entityGuid.
Use this for one policy you have the ID of. Use newrelic_alert_policies_list to search by name.
Requires a policy ID from newrelic_alert_policies_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `policy_id` (`string`, required): ID of the alert policy. Get it from newrelic_alert_policies_list. Example: 4567890

Example input: `{"account_id":1234567,"policy_id":"<policy_id>"}`

### `newrelic_dashboard_get`

Get Dashboard · Read-only

Get one New Relic dashboard with its pages and widgets.
Returns the name, description, permissions, owner, permalink, created/updated times, and every page with its widgets (id, title, visualization id, layout and rawConfiguration, which holds the widget's NRQL queries).
Use this to read what a dashboard shows or to copy its NRQL queries into newrelic_nrql_query. Use newrelic_dashboards_list to find dashboard GUIDs.
Requires a dashboard GUID from newrelic_dashboards_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `guid` (`string`, required): GUID of the dashboard. Get it from newrelic_dashboards_list. Example: MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg

Example input: `{"guid":"<guid>"}`

### `newrelic_dashboards_list`

List Dashboards · Read-only

List New Relic dashboards, optionally filtered by name.
Returns the count, a nextCursor for paging, and each dashboard's guid, name, accountId, permalink, tags, permissions, owner email, parent dashboard GUID (set on pages) and created/updated times.
Use this to find a dashboard GUID. Use newrelic_dashboard_get for a dashboard's pages and widgets.
NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor. Pass the nextCursor value from the previous response to get the next page; leave empty for the first page. Example: Mjox
- `name` (`string`, optional): Return only dashboards whose name contains this text. Example: Checkout

Example input: `{}`

### `newrelic_deployments_list`

List Deployments · Read-only

List deployment markers recorded on one New Relic entity, newest first.
Returns the entity's guid and name and, for each deployment, deploymentId, version, timestamp, user, description, changelog, commit, deepLink, deploymentType and groupId.
Use this to see what was deployed to a service and when. Use newrelic_nrql_query with FROM ChangeTrackingEvent for change events created by newrelic_change_event_create.
Requires an entity GUID from newrelic_entities_search. Without a time window New Relic returns the past hour. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `entity_guid` (`string`, required): GUID of the entity. Get it from newrelic_entities_search. Example: MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg
- `end_time` (`integer`, optional): End of the time window in Unix epoch milliseconds. Example: 1767312000000
- `limit` (`integer`, optional): Maximum number of deployments to return (New Relic's default is 100, maximum 2000). Example: 50
- `query` (`string`, optional): Filter on deployment attributes using NRQL WHERE syntax, without the WHERE keyword. Example: user = 'jane@example.com' AND deploymentType IN ('CANARY')
- `start_time` (`integer`, optional): Start of the time window in Unix epoch milliseconds. Example: 1767225600000

Example input: `{"entity_guid":"<entity_guid>"}`

### `newrelic_entities_search`

Search Entities · Read-only

Search New Relic entities (services, hosts, browser and mobile apps, monitors, dashboards, workloads) with an entity search query.
Returns the total count, a nextCursor for paging, and for each entity its guid, name, entityType, domain, type, accountId, alertSeverity, reporting flag, permalink and tags.
Use this to find an entity's GUID by name, domain, type or tag. Use newrelic_entity_get for one entity's details, newrelic_dashboards_list for dashboards and newrelic_synthetic_monitors_list for monitors.
NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `query` (`string`, required): Entity search query in New Relic's entity search syntax. Filter on name, domain (APM, BROWSER, INFRA, MOBILE, SYNTH, EXT), type, accountId, reporting, alertSeverity or tags.<key>; use = , LIKE, IN, AND and OR. Example: name LIKE 'checkout' AND domain = 'APM'
- `cursor` (`string`, optional): Pagination cursor. Pass the nextCursor value from the previous response to get the next page; leave empty for the first page. Example: Mjox

Example input: `{"query":"<query>"}`

### `newrelic_entity_get`

Get Entity · Read-only

Get one New Relic entity by its GUID.
Returns its name, entityType, domain, type, accountId, alertSeverity, reporting flag, permalink and tags; APM services also return applicationId, language and an apmSummary (apdex, error rate, response time, throughput, host and instance counts), and synthetic monitors return their type, URL, period and success rate.
Use this for one entity you already have a GUID for. Use newrelic_entities_search to find GUIDs.
Requires an entity GUID from newrelic_entities_search. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `guid` (`string`, required): GUID of the entity. Get it from newrelic_entities_search. Example: MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg

Example input: `{"guid":"<guid>"}`

### `newrelic_nrql_query`

Run NRQL Query · Read-only

Run a NRQL query against one New Relic account and return the results.
Returns the results rows (one object per row or facet) and metadata with the event types, facets, messages and resolved time window.
Use this for any telemetry question: metrics, events, logs, traces, deployments (FROM Deployment) or change events (FROM ChangeTrackingEvent). Use newrelic_entities_search to find services and hosts by name instead.
Requires an account_id from newrelic_accounts_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `query` (`string`, required): The NRQL query to run. Include a SINCE clause to set the time range (the default is the last hour). Example: SELECT count(*) FROM Transaction WHERE appName = 'checkout' FACET name SINCE 1 hour ago
- `timeout` (`integer`, optional): Seconds to wait for the query before New Relic stops it. Leave empty to use New Relic's default (5 seconds); raise it for heavy queries over long time ranges. Example: 30

Example input: `{"account_id":1234567,"query":"<query>"}`

### `newrelic_synthetic_monitors_list`

List Synthetic Monitors · Read-only

List New Relic synthetic monitors, optionally filtered by name.
Returns the count, a nextCursor for paging, and each monitor's guid, name, accountId, monitorId, monitorType (SIMPLE, BROWSER, SCRIPT_API, SCRIPT_BROWSER, CERT_CHECK, BROKEN_LINKS, STEP_MONITOR), monitoredUrl, period in minutes, alertSeverity, tags, and a summary with status, success rate and failing locations.
Use this to check uptime or find a monitor GUID. Use newrelic_entity_get for one monitor and newrelic_nrql_query with FROM SyntheticCheck for individual check results.
NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor. Pass the nextCursor value from the previous response to get the next page; leave empty for the first page. Example: Mjox
- `name` (`string`, optional): Return only monitors whose name contains this text. Example: homepage

Example input: `{}`

### `newrelic_alert_issue_acknowledge`

Acknowledge Alert Issue · Write

Acknowledge an open New Relic alert issue so the team sees it is handled.
Returns error (null on success) and a result with the action, accountId and issueId.
Use this when a responder takes the issue. Use newrelic_alert_issue_resolve to close the issue instead.
Requires an issueId from newrelic_alert_issues_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `issue_id` (`string`, required): ID of the issue. Get it from the issueId field of newrelic_alert_issues_list. Example: f0846f5e-fc9d-4dc0-afdd-07cce7cc6fc4

Example input: `{"account_id":1234567,"issue_id":"<issue_id>"}`

### `newrelic_alert_issue_resolve`

Resolve Alert Issue · Write

Resolve (close) an open New Relic alert issue so the team sees it is handled.
Returns error (null on success) and a result with the action, accountId and issueId.
Use this when a responder takes the issue. Use newrelic_alert_issue_acknowledge to only mark it as being worked on.
Requires an issueId from newrelic_alert_issues_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `issue_id` (`string`, required): ID of the issue. Get it from the issueId field of newrelic_alert_issues_list. Example: f0846f5e-fc9d-4dc0-afdd-07cce7cc6fc4

Example input: `{"account_id":1234567,"issue_id":"<issue_id>"}`

### `newrelic_alert_nrql_condition_static_create`

Create Static NRQL Alert Condition · Write

Create a static-threshold NRQL alert condition in an alert policy.
Returns the new condition's id, name, enabled flag, policyId, NRQL query, thresholds and signal settings.
Use this to alert when a NRQL value crosses a fixed number (for example error count above 5). Use newrelic_alert_nrql_condition_static_update to change an existing condition.
Requires a policy ID from newrelic_alert_policies_list or newrelic_alert_policy_create. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `name` (`string`, required): Name of the condition. Example: High error count
- `policy_id` (`string`, required): ID of the alert policy to add the condition to. Get it from newrelic_alert_policies_list or newrelic_alert_policy_create. Example: 4567890
- `query` (`string`, required): NRQL query that produces the value to evaluate. Do not add SINCE or UNTIL; the condition evaluates in its own aggregation windows. Example: SELECT count(*) FROM TransactionError WHERE appName = 'checkout'
- `terms` (`array`, required): Thresholds that open incidents. Each term needs a priority (CRITICAL or WARNING), an operator (ABOVE, ABOVE_OR_EQUALS, BELOW, BELOW_OR_EQUALS, EQUALS or NOT_EQUALS), a threshold number, a thresholdDuration in seconds (a multiple of the aggregation window, 60 to 86400) and thresholdOccurrences (ALL or AT_LEAST_ONCE). Example: [{"priority": "CRITICAL", "operator": "ABOVE", "threshold": 5, "thresholdDuration": 300, "thresholdOccurrences": "ALL"}]
- `aggregation_window` (`integer`, optional): Length in seconds of each window that data is aggregated into before evaluation (30 to 900). Leave empty for New Relic's default of 60. Example: 60
- `description` (`string`, optional): Text added to incident notifications for this condition. Example: Checkout is throwing errors; see the runbook.
- `enabled` (`boolean`, optional): Whether the condition starts evaluating as soon as it is created. Defaults to true. Example: true. Default: `true`.
- `runbook_url` (`string`, optional): URL of a runbook for responders. Example: https://wiki.example.com/runbooks/checkout-errors
- `violation_time_limit_seconds` (`integer`, optional): Seconds after which an open incident from this condition is closed automatically (300 to 2592000). Leave empty for New Relic's default of 86400 (24 hours). Example: 86400

Example input: `{"account_id":1234567,"name":"<name>","policy_id":"<policy_id>","query":"<query>","terms":[{"priority":"CRITICAL","operator":"ABOVE","threshold":5,"thresholdDuration":300,"thresholdOccurrences":"ALL"}]}`

### `newrelic_alert_nrql_condition_static_update`

Update Static NRQL Alert Condition · Write

Update a static-threshold NRQL alert condition. Only the fields you provide change.
Returns the updated condition's id, name, enabled flag, policyId, NRQL query, thresholds and signal settings.
Use this to enable or disable a condition, or change its name, query, thresholds, description or runbook. Use newrelic_alert_nrql_condition_get first to read the current settings.
Requires the ID of a STATIC condition from newrelic_alert_nrql_conditions_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `condition_id` (`string`, required): ID of the alert condition. Get it from newrelic_alert_nrql_conditions_list. Example: 34567890
- `description` (`string`, optional): New text added to incident notifications. Leave empty to keep the current text. Example: Checkout is throwing errors; see the runbook.
- `enabled` (`boolean`, optional): Set to false to disable the condition or true to enable it. Leave empty to keep the current state. Example: false
- `name` (`string`, optional): New name for the condition. Leave empty to keep the current name. Example: High error count (checkout)
- `query` (`string`, optional): New NRQL query for the condition, without SINCE or UNTIL. Leave empty to keep the current query. Example: SELECT count(*) FROM TransactionError WHERE appName = 'checkout'
- `runbook_url` (`string`, optional): New runbook URL. Leave empty to keep the current one. Example: https://wiki.example.com/runbooks/checkout-errors
- `terms` (`array`, optional): Thresholds that open incidents. Each term needs a priority (CRITICAL or WARNING), an operator (ABOVE, ABOVE_OR_EQUALS, BELOW, BELOW_OR_EQUALS, EQUALS or NOT_EQUALS), a threshold number, a thresholdDuration in seconds (a multiple of the aggregation window, 60 to 86400) and thresholdOccurrences (ALL or AT_LEAST_ONCE). Example: [{"priority": "CRITICAL", "operator": "ABOVE", "threshold": 5, "thresholdDuration": 300, "thresholdOccurrences": "ALL"}] The list replaces all current thresholds; leave empty to keep them.

Example input: `{"account_id":1234567,"condition_id":"<condition_id>"}`

### `newrelic_alert_policy_create`

Create Alert Policy · Write

Create an alert policy in a New Relic account.
Returns the new policy's id, name, incidentPreference, accountId and entityGuid.
Use this before adding conditions with newrelic_alert_nrql_condition_static_create. Use newrelic_alert_policy_update to rename an existing policy.
Requires an account_id from newrelic_accounts_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `incident_preference` (`string`, required): How incidents are grouped: PER_POLICY opens one incident for the whole policy, PER_CONDITION one per condition, PER_CONDITION_AND_TARGET one per condition and entity. Example: PER_CONDITION. One of: `PER_POLICY`, `PER_CONDITION`, `PER_CONDITION_AND_TARGET`.
- `name` (`string`, required): Name of the new alert policy. Example: Checkout service - production

Example input: `{"account_id":1234567,"incident_preference":"PER_POLICY","name":"<name>"}`

### `newrelic_alert_policy_update`

Update Alert Policy · Write

Update the name or incident preference of a New Relic alert policy. Fields you leave empty keep their current value.
Returns the updated policy's id, name, incidentPreference, accountId and entityGuid.
Use this to rename a policy or change how it groups incidents. Use newrelic_alert_nrql_condition_static_update to change a condition.
Requires a policy ID from newrelic_alert_policies_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `policy_id` (`string`, required): ID of the alert policy. Get it from newrelic_alert_policies_list. Example: 4567890
- `incident_preference` (`string`, optional): How incidents are grouped: PER_POLICY opens one incident for the whole policy, PER_CONDITION one per condition, PER_CONDITION_AND_TARGET one per condition and entity. Leave empty to keep the current value. Example: PER_POLICY. One of: `PER_POLICY`, `PER_CONDITION`, `PER_CONDITION_AND_TARGET`.
- `name` (`string`, optional): New name for the policy. Leave empty to keep the current name. Example: Checkout service - prod

Example input: `{"account_id":1234567,"policy_id":"<policy_id>"}`

### `newrelic_change_event_create`

Create Change Event · Write

Record a change tracking event (a deployment, feature flag change, deployment lifecycle step or custom change) on the one entity matched by an entity search query.
Returns changeTrackingEvent with changeTrackingId, category, type, description, shortDescription, groupId, timestamp, user, customAttributes and the matched entity's guid and name, plus any messages.
Use this for change events, which New Relic recommends over deployment markers. Use newrelic_deployment_create for a legacy deployment marker.
The entity search query must match exactly one entity. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `category` (`string`, required): Change category. Standard values: Deployment, Deployment Lifecycle, Feature Flag; other values need allow_custom_category. Example: Deployment
- `entity_search_query` (`string`, required): Entity search query that matches exactly one entity. Supports =, AND, IN and LIKE. Example: id = 'MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg' or domainId = '12345678' AND domain = 'APM'
- `type` (`string`, required): Change type within the category. Deployment: Basic, Blue Green, Canary, Rolling or Shadow. Feature Flag: Basic. Deployment Lifecycle: for example Build Upload or Image Push. Example: Rolling
- `allow_custom_category` (`boolean`, optional): Set to true to allow a category or type that is not one of New Relic's standard pairs. Example: true. Default: `false`.
- `changelog` (`string`, optional): Changelog for a Deployment change. Example: Fixed retry bug; added Apple Pay
- `commit` (`string`, optional): Commit SHA for a Deployment change. Example: 3f2a9c1e7b
- `custom_attributes` (`object`, optional): Extra key-value metadata; values may be strings, numbers or booleans. Example: {"region": "us-east-1", "instances": 2}
- `deep_link` (`string`, optional): Link to the release or build for a Deployment change. Example: https://github.com/acme/checkout/releases/tag/v2.4.1
- `description` (`string`, optional): Longer description of the change. Example: Rolling release of checkout 2.4.1 to all regions
- `feature_flag_id` (`string`, optional): Feature flag identifier. Required when category is Feature Flag. Example: new-checkout-flow
- `group_id` (`string`, optional): Identifier that groups related changes across entities. Example: release-2026-11-15
- `short_description` (`string`, optional): Short text shown on the change marker. Example: Checkout v2.4.1 rollout
- `timestamp` (`integer`, optional): Time of the change in Unix epoch milliseconds, within 24 hours of now. Leave empty to use the current time. Example: 1767225600000
- `user` (`string`, optional): Person or system that made the change. Example: jane@example.com
- `version` (`string`, optional): Deployed version. Required when category is Deployment. Example: v2.4.1

Example input: `{"category":"<category>","entity_search_query":"<entity_search_query>","type":"<type>"}`

### `newrelic_dashboard_create`

Create Dashboard · Write

Create a New Relic dashboard with one or more pages of NRQL widgets.
Returns entityResult with the new dashboard's guid, name, accountId, permissions and createdAt, and an errors array.
Use this to build a dashboard from NRQL queries. Use newrelic_dashboard_get to read an existing dashboard as a model.
Requires an account_id from newrelic_accounts_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `name` (`string`, required): Name of the new dashboard. Example: Checkout service overview
- `pages` (`array`, required): Pages of the dashboard. Each page has a name and a list of widgets. A widget has a title, a layout (column 1-12, row, width 1-12, height), a visualization id such as viz.line, viz.area, viz.bar, viz.billboard, viz.pie or viz.table, and a rawConfiguration whose nrqlQueries list the account IDs and NRQL query to chart. Example: [{"name": "Overview", "widgets": [{"title": "Throughput", "layout": {"column": 1, "row": 1, "width": 6, "height": 3}, "visualization": {"id": "viz.line"}, "rawConfiguration": {"nrqlQueries": [{"accountIds": [1234567], "query": "SELECT rate(count(*), 1 minute) FROM Transaction TIMESERIES"}]}}]}]
- `description` (`string`, optional): Short description of the dashboard. Example: Throughput, errors and latency for checkout
- `permissions` (`string`, optional): Who can see and edit the dashboard: PRIVATE (only you), PUBLIC_READ_ONLY (everyone in the account can view) or PUBLIC_READ_WRITE (everyone can edit). Defaults to PUBLIC_READ_WRITE. Example: PUBLIC_READ_ONLY. One of: `PRIVATE`, `PUBLIC_READ_ONLY`, `PUBLIC_READ_WRITE`. Default: `PUBLIC_READ_WRITE`.

Example input: `{"account_id":1234567,"name":"<name>","pages":[]}`

### `newrelic_deployment_create`

Create Deployment Marker · Write

Record a deployment marker on a New Relic entity (the legacy change tracking deployment, stored as a Deployment event).
Returns the deploymentId, entityGuid, version, timestamp, user, description, changelog, commit, deepLink, deploymentType and groupId.
Use this when a release goes out for one entity whose GUID you have. Use newrelic_change_event_create for other change categories (feature flags, deployment lifecycle) or to find the entity by a search query.
Requires an entity GUID from newrelic_entities_search. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `entity_guid` (`string`, required): GUID of the entity that was deployed (usually an APM service). Get it from newrelic_entities_search. Example: MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg
- `version` (`string`, required): Version that was deployed, such as a release tag or build number. Example: v2.4.1
- `changelog` (`string`, optional): Changelog text or a link to it. Example: Fixed retry bug; added Apple Pay
- `commit` (`string`, optional): Commit SHA that was deployed. Example: 3f2a9c1e7b
- `deep_link` (`string`, optional): Link to the release or CI run. Example: https://github.com/acme/checkout/releases/tag/v2.4.1
- `deployment_type` (`string`, optional): Deployment strategy: BASIC, BLUE_GREEN, CANARY, ROLLING, SHADOW or OTHER. Example: ROLLING. One of: `BASIC`, `BLUE_GREEN`, `CANARY`, `ROLLING`, `SHADOW`, `OTHER`.
- `description` (`string`, optional): Description of the deployment. Example: Release 2.4.1 with new payment flow
- `group_id` (`string`, optional): Identifier that groups related deployments across entities. Example: release-2026-11-15
- `timestamp` (`integer`, optional): Deployment time in Unix epoch milliseconds, within 24 hours of now. Leave empty to use the current time. Example: 1767225600000
- `user` (`string`, optional): Person or system that deployed. Example: jane@example.com

Example input: `{"entity_guid":"<entity_guid>","version":"<version>"}`

### `newrelic_entity_tags_add`

Add Entity Tags · Write

Add tags to a New Relic entity. Values are added to any values the tag key already has.
Returns an errors array, which is empty when the tags were added.
Use this to label services, hosts or dashboards (for example team or environment). Use newrelic_entity_tags_delete to remove tag keys.
Requires an entity GUID from newrelic_entities_search. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `guid` (`string`, required): GUID of the entity to tag. Get it from newrelic_entities_search. Example: MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg
- `tags` (`array`, required): Tags to add. Each tag has a key and a list of values. Example: [{"key": "team", "values": ["payments"]}, {"key": "environment", "values": ["production"]}]

Example input: `{"guid":"<guid>","tags":[]}`

### `newrelic_synthetic_monitor_simple_create`

Create Ping Monitor · Write

Create a simple (ping) synthetic monitor that checks a URL on a schedule from public locations.
Returns errors and the new monitor's guid, id, name, uri, period, status, locations and advanced options.
Use this to add uptime checks for a URL. Use newrelic_synthetic_monitor_simple_update to change or pause an existing ping monitor.
Requires an account_id from newrelic_accounts_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `name` (`string`, required): Name of the monitor. Example: Homepage uptime
- `period` (`string`, required): How often the monitor runs. Example: EVERY_5_MINUTES. One of: `EVERY_MINUTE`, `EVERY_5_MINUTES`, `EVERY_10_MINUTES`, `EVERY_15_MINUTES`, `EVERY_30_MINUTES`, `EVERY_HOUR`, `EVERY_6_HOURS`, `EVERY_12_HOURS`, `EVERY_DAY`.
- `public_locations` (`array`, required): Public location codes the monitor runs from. Example: ["AWS_US_EAST_1", "AWS_EU_WEST_1"]
- `uri` (`string`, required): URL the monitor requests. Example: https://www.example.com
- `response_validation_text` (`string`, optional): Text that must appear in the response body for the check to pass. Example: Welcome
- `status` (`string`, optional): ENABLED to start checking now, DISABLED to create it paused. Defaults to ENABLED. Example: ENABLED. One of: `ENABLED`, `DISABLED`. Default: `ENABLED`.
- `tags` (`array`, optional): Tags to add to the monitor. Example: [{"key": "team", "values": ["web"]}]
- `verify_ssl` (`boolean`, optional): Whether to validate the TLS certificate. Example: true

Example input: `{"account_id":1234567,"name":"<name>","period":"EVERY_MINUTE","public_locations":["AWS_US_EAST_1","AWS_EU_WEST_1"],"uri":"<uri>"}`

### `newrelic_synthetic_monitor_simple_update`

Update Ping Monitor · Write

Update a simple (ping) synthetic monitor. Only the fields you provide change.
Returns errors and the monitor's guid, id, name, uri, period, status, locations and advanced options.
Use this to pause or resume a ping monitor (status), or change its URL, name, frequency or locations. Use newrelic_synthetic_monitor_delete to remove it.
Requires the GUID of a SIMPLE monitor from newrelic_synthetic_monitors_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `guid` (`string`, required): GUID of the synthetic monitor. Get it from newrelic_synthetic_monitors_list. Example: MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg
- `name` (`string`, optional): New name for the monitor. Example: Homepage uptime (prod)
- `period` (`string`, optional): New frequency. Example: EVERY_15_MINUTES. One of: `EVERY_MINUTE`, `EVERY_5_MINUTES`, `EVERY_10_MINUTES`, `EVERY_15_MINUTES`, `EVERY_30_MINUTES`, `EVERY_HOUR`, `EVERY_6_HOURS`, `EVERY_12_HOURS`, `EVERY_DAY`.
- `public_locations` (`array`, optional): Public location codes the monitor runs from. Example: ["AWS_US_EAST_1", "AWS_EU_WEST_1"]
- `status` (`string`, optional): ENABLED to run the monitor, DISABLED to pause it. Leave empty to keep the current status. Example: DISABLED. One of: `ENABLED`, `DISABLED`.
- `uri` (`string`, optional): New URL to check. Example: https://www.example.com/health

Example input: `{"guid":"<guid>"}`

### `newrelic_alert_condition_delete`

Delete Alert Condition · Destructive

Delete one alert condition (NRQL or any other type) from its policy. This cannot be undone.
Returns the id of the deleted condition.
Use this to remove one condition. Use newrelic_alert_nrql_condition_static_update with enabled false to pause it instead, or newrelic_alert_policy_delete to remove a whole policy.
Requires a condition ID from newrelic_alert_nrql_conditions_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `condition_id` (`string`, required): ID of the alert condition. Get it from newrelic_alert_nrql_conditions_list. Example: 34567890

Example input: `{"account_id":1234567,"condition_id":"<condition_id>"}`

### `newrelic_alert_policy_delete`

Delete Alert Policy · Destructive

Delete a New Relic alert policy and every condition in it. This cannot be undone.
Returns the id of the deleted policy.
Use this only to remove a policy completely. Use newrelic_alert_condition_delete to remove one condition.
Requires a policy ID from newrelic_alert_policies_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`integer`, required): Numeric New Relic account ID that owns the data. Get it from newrelic_accounts_list, or from the account picker in New Relic. Example: 1234567
- `policy_id` (`string`, required): ID of the alert policy. Get it from newrelic_alert_policies_list. Example: 4567890

Example input: `{"account_id":1234567,"policy_id":"<policy_id>"}`

### `newrelic_entity_tags_delete`

Delete Entity Tags · Destructive

Remove tag keys, with all their values, from a New Relic entity.
Returns an errors array, which is empty when the tags were removed.
Use this to clean up tags. Use newrelic_entity_tags_add to add tags.
Requires an entity GUID from newrelic_entities_search. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `guid` (`string`, required): GUID of the entity. Get it from newrelic_entities_search. Example: MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg
- `tag_keys` (`array`, required): Tag keys to remove from the entity. All values of each key are removed. Example: ["team", "owner"]

Example input: `{"guid":"<guid>","tag_keys":["team","owner"]}`

### `newrelic_synthetic_monitor_delete`

Delete Synthetic Monitor · Destructive

Delete a synthetic monitor of any type. This cannot be undone.
Returns deletedGuid, the GUID of the removed monitor.
Use this to remove a monitor for good. Use newrelic_synthetic_monitor_simple_update with status DISABLED to pause a ping monitor instead.
Requires a monitor GUID from newrelic_synthetic_monitors_list. NerdGraph answers HTTP 200 even when the operation fails, so check the errors array in the response.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `guid` (`string`, required): GUID of the synthetic monitor. Get it from newrelic_synthetic_monitors_list. Example: MXxBUE18QVBQTElDQVRJT058MTIzNDU2Nzg

Example input: `{"guid":"<guid>"}`

## Related

Other Monitoring connectors ([all 48](/agentkit/connectors/?category=monitoring)):

- [Datadog](/agentkit/connectors/datadog/): Scalekit connector, API key, 105 tools
- [1Password](/agentkit/connectors/onepassword/): Scalekit connector, Access token, 4 tools
- [Malwarebytes](/agentkit/connectors/malwarebytes/): Scalekit connector, Service account, 74 tools


---

## More Scalekit documentation

| Resource | What it contains | When to use it |
|----------|-----------------|----------------|
| [/llms.txt](/llms.txt) | Structured index with routing hints per product area | Start here — find which documentation set covers your topic before loading full content |
| [/llms-full.txt](/llms-full.txt) | Complete documentation for all Scalekit products in one file | Use when you need exhaustive context across multiple products or when the topic spans several areas |
| [sitemap-0.xml](https://docs.scalekit.com/sitemap-0.xml) | Full URL list of every documentation page | Use to discover specific page URLs you can fetch for targeted, page-level answers |
