API Reference

Workers

Copy Markdown

Open in Claude

Open in ChatGPT

Open in Cursor

---

Copy Markdown

View as Markdown

Beta

BetaWorkers

List Workers

GET/accounts/{account_id}/workers/workers

Get Worker

GET/accounts/{account_id}/workers/workers/{worker_id}

Create Worker

POST/accounts/{account_id}/workers/workers

Update Worker

PUT/accounts/{account_id}/workers/workers/{worker_id}

Edit Worker

PATCH/accounts/{account_id}/workers/workers/{worker_id}

Delete Worker

DELETE/accounts/{account_id}/workers/workers/{worker_id}

ModelsExpand Collapse

Worker object { id, created_on, logpush, 8 more }

id: string

Immutable ID of the Worker.

created_on: string

When the Worker was created.

formatdate-time

logpush: boolean

Whether logpush is enabled for the Worker.

name: string

Name of the Worker.

observability: object { enabled, head_sampling_rate, logs, traces }

Observability settings for the Worker.

enabled: optional boolean

Whether observability is enabled for the Worker.

head_sampling_rate: optional number

The sampling rate for observability. From 0 to 1 (1 = 100%, 0.1 = 10%).

logs: optional object { destinations, enabled, head_sampling_rate, 2 more }

Log settings for the Worker.

destinations: optional array of string

A list of destinations where logs will be exported to.

enabled: optional boolean

Whether logs are enabled for the Worker.

head_sampling_rate: optional number

The sampling rate for logs. From 0 to 1 (1 = 100%, 0.1 = 10%).

invocation_logs: optional boolean

Whether invocation logs are enabled for the Worker.

persist: optional boolean

Whether log persistence is enabled for the Worker.

traces: optional object { destinations, enabled, head_sampling_rate, 2 more }

Trace settings for the Worker.

destinations: optional array of string

A list of destinations where traces will be exported to.

enabled: optional boolean

Whether traces are enabled for the Worker.

head_sampling_rate: optional number

The sampling rate for traces. From 0 to 1 (1 = 100%, 0.1 = 10%).

persist: optional boolean

Whether trace persistence is enabled for the Worker.

propagation_policy: optional "authenticated" or "accept"

Controls how inbound trace context (traceparent/tracestate) headers on incoming requests are handled. “authenticated” (default) honors inbound trace context only when accompanied by a valid trace auth token. “accept” unconditionally accepts inbound trace context. Requires the trace propagation feature to be enabled.

One of the following:

"authenticated"

"accept"

references: object { dispatch_namespace_outbounds, domains, durable_objects, 2 more }

Other resources that reference the Worker and depend on it existing.

dispatch_namespace_outbounds: array of object { namespace_id, namespace_name, worker_id, worker_name }

Other Workers that reference the Worker as an outbound for a dispatch namespace.

namespace_id: string

ID of the dispatch namespace.

namespace_name: string

Name of the dispatch namespace.

worker_id: string

ID of the Worker using the dispatch namespace.

worker_name: string

Name of the Worker using the dispatch namespace.

domains: array of object { id, certificate_id, hostname, 2 more }

Custom domains connected to the Worker.

id: string

ID of the custom domain.

certificate_id: string

ID of the TLS certificate issued for the custom domain.

hostname: string

Full hostname of the custom domain, including the zone name.

zone_id: string

ID of the zone.

zone_name: string

Name of the zone.

durable_objects: array of object { namespace_id, namespace_name, worker_id, worker_name }

Other Workers that reference Durable Object classes implemented by the Worker.

namespace_id: string

ID of the Durable Object namespace being used.

namespace_name: string

Name of the Durable Object namespace being used.

worker_id: string

ID of the Worker using the Durable Object implementation.

worker_name: string

Name of the Worker using the Durable Object implementation.

queues: array of object { queue_consumer_id, queue_id, queue_name }

Queues that send messages to the Worker.

queue_consumer_id: string

ID of the queue consumer configuration.

queue_id: string

ID of the queue.

queue_name: string

Name of the queue.

workers: array of object { id, name }

Other Workers that reference the Worker using service bindings.

id: string

ID of the referencing Worker.

name: string

Name of the referencing Worker.

subdomain: object { enabled, previews_enabled }

Subdomain settings for the Worker.

enabled: optional boolean

Whether the *.workers.dev subdomain is enabled for the Worker.

previews_enabled: optional boolean

Whether preview URLs are enabled for the Worker.

tags: array of string

Tags associated with the Worker.

tail_consumers: array of object { name }

Other Workers that should consume logs from the Worker.

name: string

Name of the consumer Worker.

updated_on: string

When the Worker was most recently updated.

formatdate-time

deployed_on: optional string

When the Worker’s most recent deployment was created. null if the Worker has never been deployed.

formatdate-time

WorkerDeleteResponse object { errors, messages, success }

errors: array of object { code, message, documentation_url, source }

code: number

minimum1000

message: string

documentation_url: optional string

source: optional object { pointer }

pointer: optional string

messages: array of object { code, message, documentation_url, source }

code: number

minimum1000

message: string

documentation_url: optional string

source: optional object { pointer }

pointer: optional string

success: true

Whether the API call was successful.

BetaWorkersVersions

List Versions

GET/accounts/{account_id}/workers/workers/{worker_id}/versions

Get Version

GET/accounts/{account_id}/workers/workers/{worker_id}/versions/{version_id}

Create Version

POST/accounts/{account_id}/workers/workers/{worker_id}/versions

Delete Version

DELETE/accounts/{account_id}/workers/workers/{worker_id}/versions/{version_id}

ModelsExpand Collapse

Version object { id, created_on, number, 20 more }

id: string

Version identifier.

formatuuid

created_on: string

When the version was created.

formatdate-time

number: number

The integer version number, starting from one.

urls: array of string

All routable URLs that always point to this version. Does not include alias URLs, since aliases can be updated to point to a different version.

annotations: optional object { "workers/message", "workers/tag", "workers/triggered_by" }

Metadata about the version.

"workers/message": optional string

Human-readable message about the version. Truncated to 1000 bytes if longer.

maxLength1000

"workers/tag": optional string

User-provided identifier for the version. Maximum 100 bytes.

maxLength100

"workers/triggered_by": optional string

Operation that triggered the creation of the version.

assets: optional object { config, jwt }

Configuration for assets within a Worker.

_headers and _redirects files should be included as modules named _headers and _redirects with content type text/plain.

config: optional object { html_handling, not_found_handling, run_worker_first }

Configuration for assets within a Worker.

html_handling: optional "auto-trailing-slash" or "force-trailing-slash" or "drop-trailing-slash" or "none"

Determines the redirects and rewrites of requests for HTML content.

One of the following:

"auto-trailing-slash"

"force-trailing-slash"

"drop-trailing-slash"

"none"

not_found_handling: optional "none" or "404-page" or "single-page-application"

Determines the response when a request does not match a static asset, and there is no Worker script.

One of the following:

"none"

"404-page"

"single-page-application"

run_worker_first: optional array of string or boolean

Contains a list path rules to control routing to either the Worker or assets. Glob (*) and negative (!) rules are supported. Rules must start with either ’/’ or ’!/’. At least one non-negative rule must be provided, and negative rules have higher precedence than non-negative rules.

One of the following:

array of string

Contains a list path rules to control routing to either the Worker or assets. Glob (*) and negative (!) rules are supported. Rules must start with either ’/’ or ’!/’. At least one non-negative rule must be provided, and negative rules have higher precedence than non-negative rules.

boolean

Enables routing to always invoke the Worker script ahead of all requests. When true, this is equivalent to ["/*"] in the string array version of this field.

jwt: optional string

Token provided upon successful upload of all files from a registered manifest.

bindings: optional array of object { name, type } or object { instance_name, name, type, namespace } or object { name, namespace, type } or 32 more

List of bindings attached to a Worker. You can find more about bindings on our docs: https://developers.cloudflare.com/workers/configuration/multipart-upload-metadata/#bindings.

One of the following:

AI object { name, type }

name: string

A JavaScript variable name for the binding.

type: "ai"

The kind of resource that the binding provides.

AISearch object { instance_name, name, type, namespace }

instance_name: string

The user-chosen instance name. Must exist at deploy time. The worker can search, chat, update, and manage items/jobs on this instance.

name: string

A JavaScript variable name for the binding.

type: "ai_search"

The kind of resource that the binding provides.

namespace: optional string

The namespace the instance belongs to. Defaults to “default” if omitted. Customers who don’t use namespaces can simply omit this field.

AISearchNamespace object { name, namespace, type }

name: string

A JavaScript variable name for the binding.

namespace: string

The user-chosen namespace name. Must exist before deploy — Wrangler handles auto-creation on deploy failure (R2 bucket pattern). The “default” namespace is auto-created by config-api for new accounts. Grants full access (CRUD + search + chat) to all instances within the namespace.

type: "ai_search_namespace"

The kind of resource that the binding provides.

AnalyticsEngine object { dataset, name, type }

dataset: string

The name of the dataset to bind to.

name: string

A JavaScript variable name for the binding.

type: "analytics_engine"

The kind of resource that the binding provides.

Assets object { name, type }

name: string

A JavaScript variable name for the binding.

type: "assets"

The kind of resource that the binding provides.

Browser object { name, type }

name: string

A JavaScript variable name for the binding.

type: "browser"

The kind of resource that the binding provides.

D1 object { database_id, name, type, id }

database_id: string

Identifier of the D1 database to bind to.

name: string

A JavaScript variable name for the binding.

type: "d1"

The kind of resource that the binding provides.

Deprecatedid: optional string

This property has been renamed to database_id.

Identifier of the D1 database to bind to.

DataBlob object { name, part, type }

name: string

A JavaScript variable name for the binding.

part: string

The name of the file containing the data content. Only accepted for service worker syntax Workers.

Deprecatedtype: "data_blob"

The kind of resource that the binding provides.

DispatchNamespace object { name, namespace, type, outbound }

name: string

A JavaScript variable name for the binding.

namespace: string

The name of the dispatch namespace.

type: "dispatch_namespace"

The kind of resource that the binding provides.

outbound: optional object { params, worker }

Outbound worker.

params: optional array of object { name }

Pass information from the Dispatch Worker to the Outbound Worker through the parameters.

name: string

Name of the parameter.

worker: optional object { entrypoint, environment, service }

Outbound worker.

entrypoint: optional string

Entrypoint to invoke on the outbound worker.

environment: optional string

Environment of the outbound worker.

service: optional string

Name of the outbound worker.

DurableObjectNamespace object { name, type, class_name, 4 more }

name: string

A JavaScript variable name for the binding.

type: "durable_object_namespace"

The kind of resource that the binding provides.

class_name: optional string

The exported class name of the Durable Object.

dispatch_namespace: optional string

The dispatch namespace the Durable Object script belongs to.

environment: optional string

The environment of the script_name to bind to.

namespace_id: optional string

Namespace identifier tag.

maxLength32

script_name: optional string

The script where the Durable Object is defined, if it is external to this Worker.

Hyperdrive object { id, name, type }

id: string

Identifier of the Hyperdrive connection to bind to.

name: string

A JavaScript variable name for the binding.

type: "hyperdrive"

The kind of resource that the binding provides.

Inherit object { name, type, old_name, version_id }

name: string

The name of the inherited binding.

type: "inherit"

The kind of resource that the binding provides.

old_name: optional string

The old name of the inherited binding. If set, the binding will be renamed from old_name to name in the new version. If not set, the binding will keep the same name between versions.

version_id: optional string

Identifier for the version to inherit the binding from, which can be the version ID or the literal “latest” to inherit from the latest version. Defaults to inheriting the binding from the latest version.

Images object { name, type }

name: string

A JavaScript variable name for the binding.

type: "images"

The kind of resource that the binding provides.

Json object { json, name, type }

json: unknown

JSON data to use.

name: string

A JavaScript variable name for the binding.

type: "json"

The kind of resource that the binding provides.

KVNamespace object { name, namespace_id, type }

name: string

A JavaScript variable name for the binding.

namespace_id: string

Namespace identifier tag.

maxLength32

type: "kv_namespace"

The kind of resource that the binding provides.

Media object { name, type }

name: string

A JavaScript variable name for the binding.

type: "media"

The kind of resource that the binding provides.

MTLSCertificate object { certificate_id, name, type }

certificate_id: string

Identifier of the certificate to bind to.

name: string

A JavaScript variable name for the binding.

type: "mtls_certificate"

The kind of resource that the binding provides.

PlainText object { name, text, type }

name: string

A JavaScript variable name for the binding.

text: string

The text value to use.

type: "plain_text"

The kind of resource that the binding provides.

Pipelines object { name, pipeline, type }

name: string

A JavaScript variable name for the binding.

pipeline: string

Name of the Pipeline to bind to.

type: "pipelines"

The kind of resource that the binding provides.

Queue object { name, queue_name, type }

name: string

A JavaScript variable name for the binding.

queue_name: string

Name of the Queue to bind to.

type: "queue"

The kind of resource that the binding provides.

Ratelimit object { name, namespace_id, simple, type }

name: string

A JavaScript variable name for the binding.

namespace_id: string

Identifier of the rate limit namespace to bind to.

simple: object { limit, period, mitigation_timeout }

The rate limit configuration.

limit: number

The limit (requests per period).

period: number

The period in seconds.

mitigation_timeout: optional number

Duration in seconds to apply the mitigation action after the rate limit is exceeded. Valid values are 0 (disabled), 10, or multiples of 60 up to 86400. Must be greater than or equal to the period when non-zero.

type: "ratelimit"

The kind of resource that the binding provides.

R2Bucket object { bucket_name, name, type, jurisdiction }

bucket_name: string

R2 bucket to bind to.

name: string

A JavaScript variable name for the binding.

type: "r2_bucket"

The kind of resource that the binding provides.

jurisdiction: optional "eu" or "fedramp" or "fedramp-high"

The jurisdiction of the R2 bucket.

One of the following:

"eu"

"fedramp"

"fedramp-high"

SecretText object { name, text, type }

name: string

A JavaScript variable name for the binding.

text: string

The secret value to use.

type: "secret_text"

The kind of resource that the binding provides.

SendEmail object { name, type, allowed_destination_addresses, 2 more }

name: string

A JavaScript variable name for the binding.

type: "send_email"

The kind of resource that the binding provides.

allowed_destination_addresses: optional array of string

List of allowed destination addresses.

allowed_sender_addresses: optional array of string

List of allowed sender addresses.

destination_address: optional string

Destination address for the email.

formatemail

Service object { name, service, type, 2 more }

name: string

A JavaScript variable name for the binding.

service: string

Name of Worker to bind to.

type: "service"

The kind of resource that the binding provides.

entrypoint: optional string

Entrypoint to invoke on the target Worker.

environment: optional string

Optional environment if the Worker utilizes one.

TextBlob object { name, part, type }

name: string

A JavaScript variable name for the binding.

part: string

The name of the file containing the text content. Only accepted for service worker syntax Workers.

Deprecatedtype: "text_blob"

The kind of resource that the binding provides.

Vectorize object { index_name, name, type }

index_name: string

Name of the Vectorize index to bind to.

name: string

A JavaScript variable name for the binding.

type: "vectorize"

The kind of resource that the binding provides.

VersionMetadata object { name, type }

name: string

A JavaScript variable name for the binding.

type: "version_metadata"

The kind of resource that the binding provides.

SecretsStoreSecret object { name, secret_name, store_id, type }

name: string

A JavaScript variable name for the binding.

secret_name: string

Name of the secret in the store.

store_id: string

ID of the store containing the secret.

type: "secrets_store_secret"

The kind of resource that the binding provides.

Flagship object { app_id, name, type }

app_id: string

ID of the Flagship app to bind to for feature flag evaluation.

maxLength128

name: string

A JavaScript variable name for the binding.

type: "flagship"

The kind of resource that the binding provides.

SecretKey object { algorithm, format, name, 4 more }

algorithm: unknown

Algorithm-specific key parameters. Learn more.

format: "raw" or "pkcs8" or "spki" or "jwk"

Data format of the key. Learn more.

One of the following:

"raw"

"pkcs8"

"spki"

"jwk"

name: string

A JavaScript variable name for the binding.

type: "secret_key"

The kind of resource that the binding provides.

usages: array of "encrypt" or "decrypt" or "sign" or 5 more

Allowed operations with the key. Learn more.

One of the following:

"encrypt"

"decrypt"

"sign"

"verify"

"deriveKey"

"deriveBits"

"wrapKey"

"unwrapKey"

key_base64: optional string

Base64-encoded key data. Required if format is “raw”, “pkcs8”, or “spki”.

key_jwk: optional unknown

Key data in JSON Web Key format. Required if format is “jwk”.

Workflow object { name, type, workflow_name, 2 more }

name: string

A JavaScript variable name for the binding.

type: "workflow"

The kind of resource that the binding provides.

workflow_name: string

Name of the Workflow to bind to.

class_name: optional string

Class name of the Workflow. Should only be provided if the Workflow belongs to this script.

script_name: optional string

Script name that contains the Workflow. If not provided, defaults to this script name.

WasmModule object { name, part, type }

name: string

A JavaScript variable name for the binding.

part: string

The name of the file containing the WebAssembly module content. Only accepted for service worker syntax Workers.

Deprecatedtype: "wasm_module"

The kind of resource that the binding provides.

VPCService object { name, service_id, type }

name: string

A JavaScript variable name for the binding.

service_id: string

Identifier of the VPC service to bind to.

type: "vpc_service"

The kind of resource that the binding provides.

VPCNetwork object { name, type, network_id, tunnel_id }

name: string

A JavaScript variable name for the binding.

type: "vpc_network"

The kind of resource that the binding provides.

network_id: optional string

Identifier of the network to bind to. Only “cf1:network” is currently supported. Mutually exclusive with tunnel_id.

tunnel_id: optional string

UUID of the Cloudflare Tunnel to bind to. Mutually exclusive with network_id.

cache_options: optional object { enabled, cross_version_cache }

Global CacheW configuration for the Worker. When caching is on, the platform provisions a cloudflare.app zone for the Worker. A type: worker entry in the exports map can override this value for a single entrypoint.

enabled: boolean

Whether caching is enabled for this Worker.

cross_version_cache: optional boolean

Whether cached responses are shared across Worker version uploads. This is independent of enabled. It can stay true while caching is off, so the preference survives turning caching off and back on.

compatibility_date: optional string

Date indicating targeted support in the Workers runtime. Backwards incompatible fixes to the runtime following this date will not affect this Worker.

compatibility_flags: optional array of string

Flags that enable or disable certain features in the Workers runtime. Used to enable upcoming features or opt in or out of specific changes not included in a compatibility_date.

containers: optional array of object { class_name }

List of containers attached to a Worker. Containers can only be attached to Durable Object classes of this Worker script.

class_name: string

Select which Durable Object class should get this container attached.

exports: optional map[object { type, cache, state } or object { storage, type, container, state } or object { state, type } or 3 more]

Declarative exports for the version, including Durable Object classes (with their storage backend) and named Worker entrypoints. On reads, tombstoned lifecycle entries are omitted, so only live exports (created and expecting-transfer) are returned. exports and migrations are mutually exclusive on upload.

One of the following:

Worker object { type, cache, state }

A named Worker entrypoint export (type: worker). Worker entrypoints are always live (state: created) and carry no storage or lifecycle fields. The optional cache block overrides the Worker’s global cache_options.enabled for this entrypoint.

type: "worker"

Marks this entry as a Worker entrypoint export.

cache: optional object { enabled }

Cache override for this entrypoint. Overrides the Worker’s global cache_options.enabled for this entrypoint only.

enabled: boolean

Whether caching is enabled for this entrypoint.

state: optional "created"

Live export. May be omitted; defaults to created.

WorkersDurableObjectExport object { storage, type, container, state }

A live Durable Object export (state: created, the default). The platform auto-provisions the namespace on first deploy, matches it on subsequent deploys, and never mutates or deletes it as a side effect of a code-only change. storage is required; renamed_to, transferred_to and transfer_from are not allowed on a live entry.

storage: "sqlite" or "legacy-kv"

Durable Object storage backend. sqlite is the recommended (and only) backend for new namespaces. legacy-kv is accepted only for a class whose namespace already exists as KV-backed; the exports flow never provisions a new legacy-kv namespace.

One of the following:

"sqlite"

"legacy-kv"

type: "durable-object"

Marks this entry as a Durable Object export.

container: optional string

Name of the container (declared in the upload’s metadata.containers) that backs this Durable Object. When set, the namespace is container-enabled. Valid only on live entries.

maxLength128

state: optional "created"

Live export. May be omitted; defaults to created.

WorkersDurableObjectDeletedExport object { state, type }

A deleted tombstone: retires the provisioned namespace for this class and all of its data. The class must be absent from the uploaded code and no other Worker in the account may bind to the namespace, otherwise the deploy is rejected. No other fields are allowed. Deletion is irreversible.

state: "deleted"

Tombstone that deletes the namespace.

type: "durable-object"

Marks this entry as a Durable Object export.

WorkersDurableObjectRenamedExport object { renamed_to, state, type }

A renamed tombstone: rewrites the provisioned namespace’s class name from this map key to renamed_to. The source class may stay in code during the rollout window (an info notice is emitted). storage, transferred_to and transfer_from are not allowed.

renamed_to: string

The destination class name. Must differ from the source class (the map key) and must be declared as a live (created) entry in the same exports map. Write-only: never present in GET responses.

maxLength128

state: "renamed"

Tombstone that renames the namespace’s class.

type: "durable-object"

Marks this entry as a Durable Object export.

WorkersDurableObjectTransferredExport object { state, transferred_to, type }

A transferred tombstone (source side of a two-phase transfer): hands ownership of the provisioned namespace to another script in the same account, named by transferred_to. The target must have already deployed a matching expecting-transfer entry. The source class may stay in code during the rollout window (an info notice is emitted). storage, renamed_to and transfer_from are not allowed.

state: "transferred"

Tombstone that transfers the namespace to another script.

transferred_to: string

The destination script name. Must be in the same account and the same dispatch-namespace context (or both non-dispatch). Cross-dispatch-namespace transfers are rejected. Write-only: never present in GET responses.

maxLength128

type: "durable-object"

Marks this entry as a Durable Object export.

WorkersDurableObjectExpectingTransferExport object { state, storage, transfer_from, 2 more }

The target side of a two-phase transfer (state: expecting-transfer). Declares that this script expects to receive a namespace for this class from the transfer_from script. This is a live entry, not a tombstone: bindings resolve through the source’s namespace until the source commits with a transferred tombstone. storage and transfer_from are required; renamed_to and transferred_to are not allowed.

state: "expecting-transfer"

Target side of a two-phase transfer.

storage: "sqlite" or "legacy-kv"

Durable Object storage backend. sqlite is the recommended (and only) backend for new namespaces. legacy-kv is accepted only for a class whose namespace already exists as KV-backed; the exports flow never provisions a new legacy-kv namespace.

One of the following:

"sqlite"

"legacy-kv"

transfer_from: string

The source script name to receive the namespace from. Must be in the same account and dispatch-namespace context. Present on reads for expecting-transfer entries.

maxLength128

type: "durable-object"

Marks this entry as a Durable Object export.

container: optional string

Name of the container (declared in the upload’s metadata.containers) that backs this Durable Object once the transfer settles. Valid only on live entries.

maxLength128

exports_reconciliation: optional object { created, deleted, info, 6 more }

Summary of the declarative exports reconciliation that ran on this upload. Populated only when the uploaded metadata included an exports block. Durable Object entries drive reconciliation; type: worker entries do not contribute to this summary.

created: array of string

Class names for which a new namespace was provisioned.

deleted: array of string

Class names whose namespace was deleted by a deleted tombstone.

info: array of object { class, message, scenario, 2 more }

Non-blocking info entries (stale tombstones, tombstone applied with class still in code). See exports_reconciliation_info.

class: string

The class name the info entry is about.

message: string

Human-readable explanation.

scenario: "code_class_not_in_exports" or "provisioned_class_missing_from_config" or "config_export_not_in_code" or 30 more

Stable, machine-readable tag identifying which reconciliation scenario produced an error, warning, or info entry. Clients may branch on this value instead of parsing message.

One of the following:

"code_class_not_in_exports"

"provisioned_class_missing_from_config"

"config_export_not_in_code"

"config_references_nonexistent_class"

"orphaned_provisioned_namespace"

"storage_type_mismatch"

"free_tier_requires_sqlite"

"invalid_export"

"tombstone_delete_class_still_in_code"

"tombstone_delete_blocked_by_external_bindings"

"tombstone_renamed_to_occupied"

"transferred_pending_not_found"

"transferred_target_missing"

"transferred_target_mismatch"

"phase_one_transfer_source_missing"

"phase_one_transfer_source_namespace_missing"

"phase_one_transfer_target_class_provisioned"

"phase_one_transfer_after_commit_mismatch"

"phase_one_transfer_duplicate"

"phase_one_transfer_target_in_dispatch_namespace"

"phase_one_transfer_source_in_dispatch_namespace"

"transferred_source_in_dispatch_namespace"

"transferred_target_in_dispatch_namespace"

"container_undeclared_reference"

"container_class_not_durable_object"

"container_wiring_inconsistent"

"container_multiple_durable_objects"

"transfer_container_parity_mismatch"

"transfer_container_parity_mismatch_on_commit"

"tombstone_class_still_in_code"

"stale_tombstone"

"transfer_receive_already_applied"

"transfer_receive_cleanup_complete"

namespace_id: optional string

The provisioned namespace the entry relates to, when applicable.

formatuuid

referencing_scripts: optional array of string

Other Workers in the account that still bind to the affected class. Advisory: while non-empty the tombstone is not yet safe to remove — redeploy these Workers with bindings re-pointed first.

removable_entries: array of string

Source class names whose tombstone entry is now stale and safe to delete from exports (no remaining referencing scripts).

renamed: array of object { from, to }

Applied renamed tombstones.

from: string

The original (source) class name.

to: string

The new class name (renamed_to).

transfer_pending: array of object { class, from }

Phase-1 transfer hints recorded on the target side.

class: string

The target-side class name awaiting transfer.

from: string

The source script the namespace will be transferred from.

transferred: array of object { class, phase, to }

Committed transferred tombstones (phase-2).

class: string

The source class name that was transferred.

phase: "committed"

The transfer phase. Currently always committed.

to: string

The destination script that now owns the namespace.

updated: array of string

Class names whose provisioned namespace was mutated in place.

warnings: array of object { class, message, scenario, namespace_id }

Non-blocking warnings. See exports_reconciliation_warning.

class: string

The class name the warning is about.

message: string

Human-readable explanation of the warning.

scenario: "code_class_not_in_exports" or "provisioned_class_missing_from_config" or "config_export_not_in_code" or 30 more

Stable, machine-readable tag identifying which reconciliation scenario produced an error, warning, or info entry. Clients may branch on this value instead of parsing message.

One of the following:

"code_class_not_in_exports"

"provisioned_class_missing_from_config"

"config_export_not_in_code"

"config_references_nonexistent_class"

"orphaned_provisioned_namespace"

"storage_type_mismatch"

"free_tier_requires_sqlite"

"invalid_export"

"tombstone_delete_class_still_in_code"

"tombstone_delete_blocked_by_external_bindings"

"tombstone_renamed_to_occupied"

"transferred_pending_not_found"

"transferred_target_missing"

"transferred_target_mismatch"

"phase_one_transfer_source_missing"

"phase_one_transfer_source_namespace_missing"

"phase_one_transfer_target_class_provisioned"

"phase_one_transfer_after_commit_mismatch"

"phase_one_transfer_duplicate"

"phase_one_transfer_target_in_dispatch_namespace"

"phase_one_transfer_source_in_dispatch_namespace"

"transferred_source_in_dispatch_namespace"

"transferred_target_in_dispatch_namespace"

"container_undeclared_reference"

"container_class_not_durable_object"

"container_wiring_inconsistent"

"container_multiple_durable_objects"

"transfer_container_parity_mismatch"

"transfer_container_parity_mismatch_on_commit"

"tombstone_class_still_in_code"

"stale_tombstone"

"transfer_receive_already_applied"

"transfer_receive_cleanup_complete"

namespace_id: optional string

The provisioned namespace the warning relates to, when applicable.

formatuuid

limits: optional object { cpu_ms, subrequests }

Resource limits enforced at runtime.

cpu_ms: optional number

CPU time limit in milliseconds.

subrequests: optional number

Subrequest limit per request.

main_module: optional string

The name of the main module in the modules array (e.g. the name of the module that exports a fetch handler).

migration_tag: optional string

Durable Object migration tag. Set when the version is deployed. Omitted if the version has not been deployed or the Worker does not use Durable Objects.

migrations: optional SingleStepMigration { deleted_classes, new_classes, new_sqlite_classes, 4 more } or object { new_tag, old_tag, steps }

Migrations for Durable Objects associated with the version. Migrations are applied when the version is deployed.

One of the following:

SingleStepMigration object { deleted_classes, new_classes, new_sqlite_classes, 4 more }

A single set of migrations to apply.

deleted_classes: optional array of string

A list of classes to delete Durable Object namespaces from.

new_classes: optional array of string

A list of classes to create Durable Object namespaces from.

new_sqlite_classes: optional array of string

A list of classes to create Durable Object namespaces with SQLite from.

new_tag: optional string

Tag to set as the latest migration tag.

old_tag: optional string

Tag used to verify against the latest migration tag for this Worker. If they don’t match, the upload is rejected.

renamed_classes: optional array of object { from, to }

A list of classes with Durable Object namespaces that were renamed.

from: optional string

to: optional string

transferred_classes: optional array of object { from, from_script, to }

A list of transfers for Durable Object namespaces from a different Worker and class to a class defined in this Worker.

from: optional string

from_script: optional string

to: optional string

WorkersMultipleStepMigrations object { new_tag, old_tag, steps }

new_tag: optional string

Tag to set as the latest migration tag.

old_tag: optional string

Tag used to verify against the latest migration tag for this Worker. If they don’t match, the upload is rejected.

steps: optional array of MigrationStep { deleted_classes, new_classes, new_sqlite_classes, 2 more }

Migrations to apply in order.

deleted_classes: optional array of string

A list of classes to delete Durable Object namespaces from.

new_classes: optional array of string

A list of classes to create Durable Object namespaces from.

new_sqlite_classes: optional array of string

A list of classes to create Durable Object namespaces with SQLite from.

renamed_classes: optional array of object { from, to }

A list of classes with Durable Object namespaces that were renamed.

from: optional string

to: optional string

transferred_classes: optional array of object { from, from_script, to }

A list of transfers for Durable Object namespaces from a different Worker and class to a class defined in this Worker.

from: optional string

from_script: optional string

to: optional string

modules: optional array of object { content_base64, content_type, name }

Code, sourcemaps, and other content used at runtime.

This includes _headers and _redirects files used to configure Static Assets. _headers and _redirects files should be included as modules named _headers and _redirects with content type text/plain.

content_base64: string

The base64-encoded module content.

formatbyte

content_type: string

The content type of the module.

name: string

The name of the module.

package_dependencies: optional array of object { installedVersion, name, packageJsonVersion }

The list of npm packages that were installed and used when this Worker version was built.

installedVersion: string

The exact version that was resolved and installed by the package manager.

name: string

The npm package name.

packageJsonVersion: string

The version constraint as written in package.json.

placement: optional object { mode } or object { region } or object { hostname } or 5 more

Configuration for Smart Placement. Specify mode=‘smart’ for Smart Placement, or one of region/hostname/host.

One of the following:

Mode object { mode }

mode: "smart"

Enables Smart Placement.

Region object { region }

region: string

Cloud region for targeted placement in format ‘provider:region’.

Hostname object { hostname }

hostname: string

HTTP hostname for targeted placement.

Host object { host }

host: string

TCP host and port for targeted placement.

object { mode, region }

mode: "targeted"

Targeted placement mode.

region: string

Cloud region for targeted placement in format ‘provider:region’.

object { hostname, mode }

hostname: string

HTTP hostname for targeted placement.

mode: "targeted"

Targeted placement mode.

object { host, mode }

host: string

TCP host and port for targeted placement.

mode: "targeted"

Targeted placement mode.

object { mode, target }

mode: "targeted"

Targeted placement mode.

target: array of object { region } or object { hostname } or object { host }

Array of placement targets (currently limited to single target).

One of the following:

Region object { region }

region: string

Cloud region in format ‘provider:region’.

Hostname object { hostname }

hostname: string

HTTP hostname for targeted placement.

Host object { host }

host: string

TCP host:port for targeted placement.

source: optional string

The client used to create the version.

startup_time_ms: optional number

Time in milliseconds spent on Worker startup.

Deprecatedusage_model: optional "standard" or "bundled" or "unbound"

Usage model for the version.

One of the following:

"standard"

"bundled"

"unbound"

VersionDeleteResponse object { errors, messages, success }

errors: array of object { code, message, documentation_url, source }

code: number

minimum1000

message: string

documentation_url: optional string

source: optional object { pointer }

pointer: optional string

messages: array of object { code, message, documentation_url, source }

code: number

minimum1000

message: string

documentation_url: optional string

source: optional object { pointer }

pointer: optional string

success: true

Whether the API call was successful.