Skip to content

Automate browser based workflows with AI

Workflow Copilot: fix dealing with output parameters (#4527)

Stanislav Novosadc4c1e8450794
diff --git a/skyvern/forge/prompts/skyvern/workflow_knowledge_base.txt b/skyvern/forge/prompts/skyvern/workflow_knowledge_base.txt
index ea1ebb053..1c93e4b52 100644
--- a/skyvern/forge/prompts/skyvern/workflow_knowledge_base.txt
+++ b/skyvern/forge/prompts/skyvern/workflow_knowledge_base.txt
@@ -634,18 +634,19 @@ label: <unique_label>
method: <GET|POST|PUT|PATCH|DELETE> # Optional: HTTP method (default: GET)
url: <https_url> # Optional: Target URL
headers: {} # Optional: HTTP headers
-body: {} # Optional: JSON body (dict)
+body: {} # Optional: JSON body (MUST be a dict, not a string)
files: {} # Optional: Multipart files mapping
timeout: 30 # Optional: Timeout in seconds
follow_redirects: true # Optional: Follow redirects
-parameter_keys: [] # Optional: Parameters used in this block
+parameter_keys: [] # Optional: Workflow parameters used (block outputs don't need to be listed)
Use Cases:
- Call third-party APIs for enrichment
- Post data to internal services
- Upload files via multipart requests
+- Send results from previous blocks to webhooks
-Example:
+Example 1 - Using workflow parameters:
workflow_definition:
version: 2
blocks:
@@ -667,26 +668,75 @@ workflow_definition:
key: customer_email
workflow_parameter_type: string
+Example 2 - Sending block output to webhook (no parameter_keys needed):
+workflow_definition:
+ version: 2
+ parameters: []
+ blocks:
+ - block_type: task_v2
+ label: get_data
+ next_block_label: send_webhook
+ prompt: "Get top 3 hacker news items"
+ url: "https://news.ycombinator.com"
+ - block_type: http_request
+ label: send_webhook
+ next_block_label: null
+ method: POST
+ url: "http://example.com/webhook"
+ body:
+ data: "{{ get_data.output }}"
+ parameter_keys: []
+
** PARAMETER TEMPLATING **
-All string fields in blocks support Jinja2 templating to reference parameters.
+All string fields in blocks support Jinja2 templating to reference parameters and block outputs.
+
+There are TWO types of references:
+
+1. WORKFLOW PARAMETERS - Reference input parameters defined in the parameters section
+ Syntax: {{ param_key }}
+ Requires: parameter_keys list must include the param_key
-Syntax (preferred):
-{{ param_key }}
+2. BLOCK OUTPUTS - Reference output from a previous block by its label
+ Syntax: {{ block_label.output }}
+ Requires: NOTHING - block outputs are automatically available, no parameter_keys needed
-Examples:
+IMPORTANT: Block outputs use the block's label directly (e.g., {{ block_1.output }}).
+
+Examples - Workflow Parameters:
* In URL:
url: "https://example.com/search?q={{ search_term }}"
+parameter_keys: [search_term]
* In goals:
navigation_goal: "Search for {{ product_name }} and filter by {{ category }}"
+parameter_keys: [product_name, category]
-* In data extraction:
-data_extraction_goal: "Extract {{ field_name }} from the results"
+Examples - Block Outputs (no parameter_keys needed):
-* Complex expressions:
-navigation_goal: "Enter {{ first_name }} {{ last_name }} in the name field"
+* Send previous block output to webhook:
+blocks:
+ - block_type: task_v2
+ label: get_data
+ prompt: "Get top 3 hacker news items"
+ ...
+ - block_type: http_request
+ label: send_webhook
+ method: POST
+ url: "http://example.com/webhook"
+ body:
+ data: "{{ get_data.output }}"
+ parameter_keys: [] # Empty - block outputs don't need to be declared
+
+* Use extraction output in next block:
+blocks:
+ - block_type: extraction
+ label: extract_items
+ ...
+ - block_type: task_v2
+ label: process_items
+ prompt: "Process these items: {{ extract_items.output }}"
* In schemas (as descriptions):
data_schema:
diff --git a/skyvern/forge/sdk/routes/workflow_copilot.py b/skyvern/forge/sdk/routes/workflow_copilot.py
index d96a759e1..283010ac9 100644
--- a/skyvern/forge/sdk/routes/workflow_copilot.py
+++ b/skyvern/forge/sdk/routes/workflow_copilot.py
@@ -204,7 +204,7 @@ async def copilot_call_llm(
if action_type == "REPLACE_WORKFLOW":
llm_workflow_yaml = action_data.get("workflow_yaml", "")
try:
- updated_workflow = await _process_workflow_yaml(
+ updated_workflow = _process_workflow_yaml(
workflow_id=chat_request.workflow_id,
workflow_permanent_id=chat_request.workflow_permanent_id,
organization_id=organization_id,
@@ -228,7 +228,7 @@ async def copilot_call_llm(
debug_run_info_text=debug_run_info_text,
error=e,
)
- updated_workflow = await _process_workflow_yaml(
+ updated_workflow = _process_workflow_yaml(
workflow_id=chat_request.workflow_id,
workflow_permanent_id=chat_request.workflow_permanent_id,
organization_id=organization_id,
@@ -298,7 +298,7 @@ async def _auto_correct_workflow_yaml(
return action_data.get("workflow_yaml", workflow_yaml)
-async def _process_workflow_yaml(
+def _process_workflow_yaml(
workflow_id: str,
workflow_permanent_id: str,
organization_id: str,