Query Cloudflare Workers Observability Logs

Query the CF Workers Observability API to retrieve structured logs from any Workers, Durable Objects, Workflows, Queues, and Cron Triggers in your Cloudflare account.

Prerequisites

Set these environment variables (e.g. in your project's .env or shell profile):

• CF_OBSERVABILITY_ACCOUNT_ID — your Cloudflare account ID
• CF_OBSERVABILITY_API_TOKEN — API token with Workers Observability read permission

API Endpoint

POST https://api.cloudflare.com/client/v4/accounts/{accountId}/workers/observability/telemetry/query
Authorization: Bearer {apiToken}
Content-Type: application/json

Request Body Format

{
  "queryId": "cc-{timestamp}",
  "timeframe": {
    "from": "<unix_ms_start>",
    "to": "<unix_ms_end>"
  },
  "view": "events",
  "limit": 50,
  "parameters": {
    "filters": [
      {"key": "<field>", "operation": "<op>", "type": "<type>", "value": "<value>"}
    ],
    "filterCombination": "and",
    "calculations": [],
    "groupBys": [],
    "needle": {"value": "<search_text>", "isRegex": false, "matchCase": false},
    "limit": 50
  }
}

Filter Operations

• String: eq, neq, includes, doesNotInclude, startsWith, regex, exists, doesNotExist
• Number: eq, neq, gt, gte, lt, lte, exists, doesNotExist

Standard Fields

Any custom fields logged via console.log({ key: value }) are also queryable as top-level fields.

Full-text Search

Use needle.value for free-text search across all fields. Useful when you don't know which field contains the value.

How to Execute

Use Bash with curl to call the API. Do NOT use WebFetch (it processes through an AI model and loses structure).

Step 1: Read credentials

Read CF_OBSERVABILITY_ACCOUNT_ID and CF_OBSERVABILITY_API_TOKEN from environment variables. If not set in the shell, search for them in project .env files:

grep -r 'CF_OBSERVABILITY_' --include='.env' --include='.env.*' . 2>/dev/null

Step 2: Build and execute query

Construct the curl command based on the user's request. Default time range: last 1 hour. Default limit: 30.

Step 3: Format output

Parse the JSON response and format as a timeline:

{timestamp} [{level}] [{scriptName}/{entrypoint}] {msg}
         {extra fields if present: error=, status=, eventType=}