fix: route email trigger path through standard info channel (#8996)

* docs(skill): document email triggers and S3 attachments

Add an "Email triggers" section to the triggers skill covering the
local-part config, the parsed_email/raw_email/email_extra_args payload,
the URL-style extras convention, where to find trigger_path (only with
a preprocessor, at event.trigger_path), and — most importantly — that
binary attachments are uploaded to the workspace S3 bucket and surface
as `{ s3: "windmill_emails/<job_id>/attachments/<filename>" }`. Scripts
must use wmill.loadS3File / wmill.load_s3_file to read them.

Also pulls EmailTrigger into the schema mappings so a real
`email_trigger.schema.yaml` is generated, and adds Email/Azure to the
trigger kinds list in the CLI agent guidance.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* chore: update ee-repo-ref for email trigger path fix

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* chore: update ee-repo-ref to 26184ab7a4aadfc529dcedf038aa08d36c7ad381

This commit updates the EE repository reference after PR #553 was merged in windmill-ee-private.

Previous ee-repo-ref: 318a46897a605dc9be3817901f35ba5a99a0a525

New ee-repo-ref: 26184ab7a4aadfc529dcedf038aa08d36c7ad381

Automated by sync-ee-ref workflow.

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>

backend/.sqlx/query-4d73cefcc1b72238b7731ab52376b849a6f712b4109a48d9f77da8dd0060ab1a.json

@@ -1,6 +1,6 @@
 {
   "db_name": "PostgreSQL",
-  "query": "\n            SELECT EXISTS(\n                SELECT 1\n                FROM email_trigger\n                WHERE\n                    ((workspaced_local_part IS TRUE AND workspace_id || '-' || local_part = $1)\n                    OR (workspaced_local_part IS FALSE AND local_part = $1))\n                    AND ($2::TEXT IS NULL OR path != $2)\n            )\n            ",
+  "query": "\n            SELECT EXISTS(\n                SELECT 1\n                FROM email_trigger\n                WHERE\n                    ((workspaced_local_part IS TRUE AND workspace_id || '-' || local_part = $1)\n                    OR (workspaced_local_part IS FALSE AND local_part = $1))\n                    AND ($2::TEXT IS NULL OR NOT (workspace_id = $3 AND path = $2))\n            )\n            ",
   "describe": {
     "columns": [
       {
@@ -11,6 +11,7 @@
     ],
     "parameters": {
       "Left": [
+        "Text",
         "Text",
         "Text"
       ]
@@ -19,5 +20,5 @@
       null
     ]
   },
-  "hash": "bb94e3105cb1bc5d90af2bc914b579a6e821c432eac38ad877d9ff362d8ab916"
+  "hash": "4d73cefcc1b72238b7731ab52376b849a6f712b4109a48d9f77da8dd0060ab1a"
 }

backend/ee-repo-ref.txt

@@ -1 +1 @@
-1790e8df54bb00fb6800b8f3a6a2bb4c9787380e
+26184ab7a4aadfc529dcedf038aa08d36c7ad381

cli/src/guidance/core.ts

@@ -37,7 +37,7 @@ When a new app needs to be created, YOU run \`wmill app new\` yourself with \`--
 
 ## Triggers
 
-You MUST use the \`triggers\` skill to configure HTTP routes, WebSocket, Kafka, NATS, SQS, MQTT, GCP, or Postgres CDC triggers.
+You MUST use the \`triggers\` skill to configure HTTP routes, WebSocket, Kafka, NATS, SQS, MQTT, GCP, Azure, Email, or Postgres CDC triggers.
 
 ## Schedules
 

cli/src/guidance/skills.gen.ts

@@ -5563,6 +5563,49 @@ Examples:
 - \`u/user/webhook.http_trigger.yaml\`
 - \`f/data/kafka_consumer.kafka_trigger.yaml\`
 - \`f/sync/postgres_cdc.postgres_trigger.yaml\`
+- \`f/inbound/orders.email_trigger.yaml\`
+
+## Email Triggers
+
+An email trigger routes incoming emails to a script or flow. Each trigger reserves a local-part: emails sent to \`<local_part>@<windmill_email_domain>\` are delivered to the configured runnable. Set \`workspaced_local_part: true\` to namespace it per workspace (the actual recipient becomes \`<workspace_id>-<local_part>@…\`); on Windmill Cloud this is required.
+
+Senders may append URL-style extras to the local-part with \`+\`: \`mytrigger+foo=bar+baz=qux@…\`. They flow through to the script as \`email_extra_args\` (see below).
+
+### Payload
+
+The runnable receives:
+
+- \`parsed_email\` — \`{ headers, text_body, html_body, attachments[] }\`. Each \`attachment\` has \`{ headers, body }\`.
+- \`raw_email\` — the raw RFC 822 message as a string, **or** an S3 object (\`{ s3: "windmill_emails/<job_id>/raw.eml" }\`) if the message exceeds 1 MiB.
+- \`email_extra_args\` (optional, only when sender appended \`+key=value\` extras) — a flat object of the parsed extras.
+
+With a preprocessor, all of the above are nested under \`event\` along with \`event.kind = "email"\` and \`event.trigger_path\` (the trigger's path). Without a preprocessor, \`trigger_path\` is **not** exposed — add a preprocessor if you need it.
+
+### Attachments are S3 objects
+
+Binary attachments are uploaded to the workspace S3 bucket and surface in \`parsed_email.attachments[i].body\` as:
+
+\`\`\`json
+{ "s3": "windmill_emails/<job_id>/attachments/<filename>" }
+\`\`\`
+
+To read the bytes inside a script, use the wmill SDK:
+
+\`\`\`ts
+// TypeScript
+import * as wmill from "windmill-client"
+const file = await wmill.loadS3File(parsed_email.attachments[0].body)
+\`\`\`
+
+\`\`\`python
+# Python
+import wmill
+data = wmill.load_s3_file(parsed_email["attachments"][0]["body"])
+\`\`\`
+
+If the workspace has no S3 resource configured (Workspace Settings → Object storage), \`body\` falls back to the string \`"configure s3 in the workspace settings to handle attachments"\`. The same applies to large \`raw_email\` bodies. Email attachment storage requires the server to be built with the \`parquet\` feature.
+
+Text/HTML/inline parts are placed inline in \`body\` as strings.
 
 ## CLI Commands
 
@@ -7180,6 +7223,71 @@ required:
 - azure_mode
 - scope_resource_id
 - subscription_name
+`,
+  "email_trigger": `type: object
+properties:
+  script_path:
+    type: string
+    description: Path to the script or flow to execute when triggered
+  permissioned_as:
+    type: string
+    description: The user or group this trigger runs as (permissioned_as)
+  is_flow:
+    type: boolean
+    description: True if script_path points to a flow, false if it points to a script
+  labels:
+    type: array
+    items:
+      type: string
+  local_part:
+    type: string
+  workspaced_local_part:
+    type: boolean
+  error_handler_path:
+    type: string
+  error_handler_args:
+    type: object
+    description: The arguments to pass to the script or flow
+  retry:
+    type: object
+    properties:
+      constant:
+        type: object
+        description: Retry with constant delay between attempts
+        properties:
+          attempts:
+            type: integer
+            description: Number of retry attempts
+          seconds:
+            type: integer
+            description: Seconds to wait between retries
+      exponential:
+        type: object
+        description: Retry with exponential backoff (delay doubles each time)
+        properties:
+          attempts:
+            type: integer
+            description: Number of retry attempts
+          multiplier:
+            type: integer
+            description: Multiplier for exponential backoff
+          seconds:
+            type: integer
+            minimum: 1
+            description: Initial delay in seconds
+          random_factor:
+            type: integer
+            minimum: 0
+            maximum: 100
+            description: Random jitter percentage (0-100) to avoid thundering herd
+      retry_if:
+        $ref: '#/components/schemas/RetryIf'
+    description: Retry configuration for failed module executions
+required:
+- script_path
+- permissioned_as
+- is_flow
+- local_part
 `,
   "gcp_trigger": `type: object
 properties:
@@ -8116,6 +8224,7 @@ export const SCHEMA_MAPPINGS: Record<string, SchemaMapping[]> = {
     { name: "SqsTrigger", schemaKey: "sqs_trigger", filePattern: "*.sqs_trigger.yaml" },
     { name: "GcpTrigger", schemaKey: "gcp_trigger", filePattern: "*.gcp_trigger.yaml" },
     { name: "AzureTrigger", schemaKey: "azure_trigger", filePattern: "*.azure_trigger.yaml" },
+    { name: "EmailTrigger", schemaKey: "email_trigger", filePattern: "*.email_trigger.yaml" },
   ],
   "schedules": [
     { name: "Schedule", schemaKey: "schedule", filePattern: "*.schedule.yaml" },

system_prompts/auto-generated/schemas/email_trigger.schema.yaml

@@ -0,0 +1,64 @@
+type: object
+properties:
+  script_path:
+    type: string
+    description: Path to the script or flow to execute when triggered
+  permissioned_as:
+    type: string
+    description: The user or group this trigger runs as (permissioned_as)
+  is_flow:
+    type: boolean
+    description: True if script_path points to a flow, false if it points to a script
+  labels:
+    type: array
+    items:
+      type: string
+  local_part:
+    type: string
+  workspaced_local_part:
+    type: boolean
+  error_handler_path:
+    type: string
+  error_handler_args:
+    type: object
+    description: The arguments to pass to the script or flow
+  retry:
+    type: object
+    properties:
+      constant:
+        type: object
+        description: Retry with constant delay between attempts
+        properties:
+          attempts:
+            type: integer
+            description: Number of retry attempts
+          seconds:
+            type: integer
+            description: Seconds to wait between retries
+      exponential:
+        type: object
+        description: Retry with exponential backoff (delay doubles each time)
+        properties:
+          attempts:
+            type: integer
+            description: Number of retry attempts
+          multiplier:
+            type: integer
+            description: Multiplier for exponential backoff
+          seconds:
+            type: integer
+            minimum: 1
+            description: Initial delay in seconds
+          random_factor:
+            type: integer
+            minimum: 0
+            maximum: 100
+            description: Random jitter percentage (0-100) to avoid thundering herd
+      retry_if:
+        $ref: '#/components/schemas/RetryIf'
+    description: Retry configuration for failed module executions
+required:
+- script_path
+- permissioned_as
+- is_flow
+- local_part

system_prompts/auto-generated/skills/triggers/SKILL.md

@@ -15,6 +15,49 @@ Examples:
 - `u/user/webhook.http_trigger.yaml`
 - `f/data/kafka_consumer.kafka_trigger.yaml`
 - `f/sync/postgres_cdc.postgres_trigger.yaml`
+- `f/inbound/orders.email_trigger.yaml`
+
+## Email Triggers
+
+An email trigger routes incoming emails to a script or flow. Each trigger reserves a local-part: emails sent to `<local_part>@<windmill_email_domain>` are delivered to the configured runnable. Set `workspaced_local_part: true` to namespace it per workspace (the actual recipient becomes `<workspace_id>-<local_part>@…`); on Windmill Cloud this is required.
+
+Senders may append URL-style extras to the local-part with `+`: `mytrigger+foo=bar+baz=qux@…`. They flow through to the script as `email_extra_args` (see below).
+
+### Payload
+
+The runnable receives:
+
+- `parsed_email` — `{ headers, text_body, html_body, attachments[] }`. Each `attachment` has `{ headers, body }`.
+- `raw_email` — the raw RFC 822 message as a string, **or** an S3 object (`{ s3: "windmill_emails/<job_id>/raw.eml" }`) if the message exceeds 1 MiB.
+- `email_extra_args` (optional, only when sender appended `+key=value` extras) — a flat object of the parsed extras.
+
+With a preprocessor, all of the above are nested under `event` along with `event.kind = "email"` and `event.trigger_path` (the trigger's path). Without a preprocessor, `trigger_path` is **not** exposed — add a preprocessor if you need it.
+
+### Attachments are S3 objects
+
+Binary attachments are uploaded to the workspace S3 bucket and surface in `parsed_email.attachments[i].body` as:
+
+```json
+{ "s3": "windmill_emails/<job_id>/attachments/<filename>" }
+```
+
+To read the bytes inside a script, use the wmill SDK:
+
+```ts
+// TypeScript
+import * as wmill from "windmill-client"
+const file = await wmill.loadS3File(parsed_email.attachments[0].body)
+```
+
+```python
+# Python
+import wmill
+data = wmill.load_s3_file(parsed_email["attachments"][0]["body"])
+```
+
+If the workspace has no S3 resource configured (Workspace Settings → Object storage), `body` falls back to the string `"configure s3 in the workspace settings to handle attachments"`. The same applies to large `raw_email` bodies. Email attachment storage requires the server to be built with the `parquet` feature.
+
+Text/HTML/inline parts are placed inline in `body` as strings.
 
 ## CLI Commands
 

system_prompts/base/triggers.md

@@ -10,6 +10,49 @@ Examples:
 - `u/user/webhook.http_trigger.yaml`
 - `f/data/kafka_consumer.kafka_trigger.yaml`
 - `f/sync/postgres_cdc.postgres_trigger.yaml`
+- `f/inbound/orders.email_trigger.yaml`
+
+## Email Triggers
+
+An email trigger routes incoming emails to a script or flow. Each trigger reserves a local-part: emails sent to `<local_part>@<windmill_email_domain>` are delivered to the configured runnable. Set `workspaced_local_part: true` to namespace it per workspace (the actual recipient becomes `<workspace_id>-<local_part>@…`); on Windmill Cloud this is required.
+
+Senders may append URL-style extras to the local-part with `+`: `mytrigger+foo=bar+baz=qux@…`. They flow through to the script as `email_extra_args` (see below).
+
+### Payload
+
+The runnable receives:
+
+- `parsed_email` — `{ headers, text_body, html_body, attachments[] }`. Each `attachment` has `{ headers, body }`.
+- `raw_email` — the raw RFC 822 message as a string, **or** an S3 object (`{ s3: "windmill_emails/<job_id>/raw.eml" }`) if the message exceeds 1 MiB.
+- `email_extra_args` (optional, only when sender appended `+key=value` extras) — a flat object of the parsed extras.
+
+With a preprocessor, all of the above are nested under `event` along with `event.kind = "email"` and `event.trigger_path` (the trigger's path). Without a preprocessor, `trigger_path` is **not** exposed — add a preprocessor if you need it.
+
+### Attachments are S3 objects
+
+Binary attachments are uploaded to the workspace S3 bucket and surface in `parsed_email.attachments[i].body` as:
+
+```json
+{ "s3": "windmill_emails/<job_id>/attachments/<filename>" }
+```
+
+To read the bytes inside a script, use the wmill SDK:
+
+```ts
+// TypeScript
+import * as wmill from "windmill-client"
+const file = await wmill.loadS3File(parsed_email.attachments[0].body)
+```
+
+```python
+# Python
+import wmill
+data = wmill.load_s3_file(parsed_email["attachments"][0]["body"])
+```
+
+If the workspace has no S3 resource configured (Workspace Settings → Object storage), `body` falls back to the string `"configure s3 in the workspace settings to handle attachments"`. The same applies to large `raw_email` bodies. Email attachment storage requires the server to be built with the `parquet` feature.
+
+Text/HTML/inline parts are placed inline in `body` as strings.
 
 ## CLI Commands
 

system_prompts/generate.py

@@ -1432,6 +1432,7 @@ SKILL_DEFINITIONS = [
             ('SqsTrigger', 'sqs_trigger'),
             ('GcpTrigger', 'gcp_trigger'),
             ('AzureTrigger', 'azure_trigger'),
+            ('EmailTrigger', 'email_trigger'),
         ],
     },
     {
@@ -1893,6 +1894,7 @@ def main():
             'SqsTrigger', 'NewSqsTrigger',
             'GcpTrigger',
             'AzureTrigger',
+            'EmailTrigger', 'NewEmailTrigger',
         ]
         for schema_name in schema_names:
             if schema_name in backend_schemas:

system_prompts/utils.py

@@ -49,6 +49,7 @@ SCHEMA_MAPPINGS = {
         ('SqsTrigger', 'sqs_trigger'),
         ('GcpTrigger', 'gcp_trigger'),
         ('AzureTrigger', 'azure_trigger'),
+        ('EmailTrigger', 'email_trigger'),
     ],
     'schedules': [
         ('Schedule', 'schedule'),