Add Skill: unifai-workflow-execution

Extracted from: https://github.com/redhat-community-ai-tools/UnifAI.git
Score: 1.0

skills/research-pipeline/SKILL.md

@@ -1,27 +1,26 @@
 ---
 name: research-pipeline
 version: 1.0.0
-description: Fetch a Wikipedia page, summarise it using an AI agent, and write the
-  summary to a file.
+description: Fetch a Wikipedia page, summarise its content using an AI agent, and
+  write the summary to a file.
 inputs:
 - URL of the Wikipedia page
 steps:
-- 'Step 1: Import necessary modules from _bootstrap (NIM_MODEL, require_nim_api_key)
-  and blacknode'
-- 'Step 2: Require NVIDIA NIM API key using `require_nim_api_key()`'
-- 'Step 3: Create a Graph instance `g`'
-- 'Step 4: Add a Literal node for the URL of the Wikipedia page (`url`)'
-- 'Step 5: Add an HTTPGet node to fetch the content from the URL (`fetcher`) and connect
-  it to the Literal node'
-- 'Step 6: Add an LLMAgent node with system prompt ''You are a technical writer. Summarise
-  the text in 3 bullet points.'' and model NIM_MODEL (`summarise`), connecting its
-  input to the output of `fetcher`'
-- 'Step 7: Add a FileWrite node to write the summary to a file named ''summary.txt''
-  (`writer`), connecting its input to the output of `summarise`'
-- 'Step 8: Cook the graph starting from the writer node and print the path where the
-  summary is written'
+- 'Step 1: Import necessary modules from `_bootstrap` and `blacknode`: Run `from _bootstrap
+  import NIM_MODEL, require_nim_api_key; import blacknode as bn`'
+- 'Step 2: Require NVIDIA NIM API key: Call `require_nim_api_key()` to ensure the
+  API key is set.'
+- 'Step 3: Create a graph instance: Initialize `g = bn.Graph()`.'
+- 'Step 4: Add nodes for URL, HTTPGet, summarisation, and file writing: `url = g.node(''Literal'',
+  value=''URL of the Wikipedia page''); fetcher = g.node(''HTTPGet''); summarise =
+  g.node(''LLMAgent'', system=''You are a technical writer. Summarise the text in
+  3 bullet points.'', model=NIM_MODEL); writer = g.node(''FileWrite'', path=''summary.txt'')`'
+- 'Step 5: Connect nodes with edges: `url.out(''value'') >> fetcher.inp(''url'');
+  fetcher.out(''text'') >> summarise.inp(''prompt''); summarise.out(''text'') >> writer.inp(''text'')`'
+- 'Step 6: Cook the graph to execute and get output: `result = g.cook(writer, ''path'');
+  print(f''Summary written to: {result}'')`'
 outputs:
-- Path to the summary file
+- Path of the summary file
 tags: []
 metadata:
   source_repo: https://github.com/temiroff/Blacknode.git
@@ -31,7 +30,7 @@ metadata:
 
 # research-pipeline
 
-Fetch a Wikipedia page, summarise it using an AI agent, and write the summary to a file.
+Fetch a Wikipedia page, summarise its content using an AI agent, and write the summary to a file.
 
 ## Setup
 
@@ -43,8 +42,8 @@ pip install anthropic>=0.25 docker>=7.1 openai>=1.0
 
 **Setup steps:**
 
-1. Ensure NVIDIA NIM API key is set in the environment or editor UI
-1. Install required dependencies using `pip install -r requirements.txt`
+1. Ensure NVIDIA NIM API key is set in the environment or editor
+1. Install required dependencies: `pip install -r requirements.txt`
 
 ## Key Files
 
@@ -52,39 +51,25 @@ pip install anthropic>=0.25 docker>=7.1 openai>=1.0
 
 ## Steps
 
-1. Step 1: Import necessary modules from _bootstrap (NIM_MODEL, require_nim_api_key) and blacknode
-2. Step 2: Require NVIDIA NIM API key using `require_nim_api_key()`
-3. Step 3: Create a Graph instance `g`
-4. Step 4: Add a Literal node for the URL of the Wikipedia page (`url`)
-5. Step 5: Add an HTTPGet node to fetch the content from the URL (`fetcher`) and connect it to the Literal node
-6. Step 6: Add an LLMAgent node with system prompt 'You are a technical writer. Summarise the text in 3 bullet points.' and model NIM_MODEL (`summarise`), connecting its input to the output of `fetcher`
-7. Step 7: Add a FileWrite node to write the summary to a file named 'summary.txt' (`writer`), connecting its input to the output of `summarise`
-8. Step 8: Cook the graph starting from the writer node and print the path where the summary is written
+1. Step 1: Import necessary modules from `_bootstrap` and `blacknode`: Run `from _bootstrap import NIM_MODEL, require_nim_api_key; import blacknode as bn`
+2. Step 2: Require NVIDIA NIM API key: Call `require_nim_api_key()` to ensure the API key is set.
+3. Step 3: Create a graph instance: Initialize `g = bn.Graph()`.
+4. Step 4: Add nodes for URL, HTTPGet, summarisation, and file writing: `url = g.node('Literal', value='URL of the Wikipedia page'); fetcher = g.node('HTTPGet'); summarise = g.node('LLMAgent', system='You are a technical writer. Summarise the text in 3 bullet points.', model=NIM_MODEL); writer = g.node('FileWrite', path='summary.txt')`
+5. Step 5: Connect nodes with edges: `url.out('value') >> fetcher.inp('url'); fetcher.out('text') >> summarise.inp('prompt'); summarise.out('text') >> writer.inp('text')`
+6. Step 6: Cook the graph to execute and get output: `result = g.cook(writer, 'path'); print(f'Summary written to: {result}')`
 
 ## Implementation Details
 
 ```python
-from _bootstrap import NIM_MODEL, require_nim_api_key
-import blacknode as bn
+from _bootstrap import NIM_MODEL, require_nim_api_key; import blacknode as bn
 ```
 
 ```python
-g = bn.Graph()
-url = g.node('Literal', value='https://en.wikipedia.org/w/api.php?action=query&prop=extracts&exintro=1&explaintext=1&titles=Houdini_(software)&format=json&formatversion=2&origin=*')
-fetcher = g.node('HTTPGet')
-summarise = g.node('LLMAgent', system='You are a technical writer. Summarise the text in 3 bullet points.', model=NIM_MODEL)
-writer = g.node('FileWrite', path='summary.txt')
+url = g.node('Literal', value='https://en.wikipedia.org/w/api.php?action=query&prop=extracts&exintro=1&explaintext=1&titles=Houdini_(software)&format=json&formatversion=2&origin=*'); fetcher = g.node('HTTPGet'); summarise = g.node('LLMAgent', system='You are a technical writer. Summarise the text in 3 bullet points.', model=NIM_MODEL); writer = g.node('FileWrite', path='summary.txt')
 ```
 
 ```python
-url.out('value') >> fetcher.inp('url')
-fetcher.out('text') >> summarise.inp('prompt')
-summarise.out('text') >> writer.inp('text')
-```
-
-```python
-result = g.cook(writer, 'path')
-print(f'Summary written to: {result}')
+url.out('value') >> fetcher.inp('url'); fetcher.out('text') >> summarise.inp('prompt'); summarise.out('text') >> writer.inp('text')
 ```
 
 ## Inputs
@@ -93,11 +78,11 @@ print(f'Summary written to: {result}')
 
 ## Outputs
 
-- Path to the summary file
+- Path of the summary file
 
 ## Failure Modes
 
-- If the URL is invalid, HTTPGet will fail; if NIM API key is missing, LLMAgent will not function properly
+- If the URL is invalid or unreachable, the HTTPGet node will fail; if the summarisation fails, the output text might be empty
 
 ## Source
 

skills/research-pipeline/examples.md

@@ -5,6 +5,6 @@
 ```python
 # How to use this skill
 # Inputs: URL of the Wikipedia page
-# Process: Step 1: Import necessary modules from _bootstrap (NIM_MODEL, require_nim_api_key) and blacknode → Step 2: Require NVIDIA NIM API key using `require_nim_api_key()` → Step 3: Create a Graph instance `g`
-# Outputs: Path to the summary file
+# Process: Step 1: Import necessary modules from `_bootstrap` and `blacknode`: Run `from _bootstrap import NIM_MODEL, require_nim_api_key; import blacknode as bn` → Step 2: Require NVIDIA NIM API key: Call `require_nim_api_key()` to ensure the API key is set. → Step 3: Create a graph instance: Initialize `g = bn.Graph()`.
+# Outputs: Path of the summary file
 ```

skills/research-pipeline/metadata.json

@@ -1,28 +1,26 @@
 {
   "name": "research-pipeline",
   "version": "1.0.0",
-  "goal": "Fetch a Wikipedia page, summarise it using an AI agent, and write the summary to a file.",
+  "goal": "Fetch a Wikipedia page, summarise its content using an AI agent, and write the summary to a file.",
   "inputs": [
     "URL of the Wikipedia page"
   ],
   "steps": [
-    "Step 1: Import necessary modules from _bootstrap (NIM_MODEL, require_nim_api_key) and blacknode",
-    "Step 2: Require NVIDIA NIM API key using `require_nim_api_key()`",
-    "Step 3: Create a Graph instance `g`",
-    "Step 4: Add a Literal node for the URL of the Wikipedia page (`url`)",
-    "Step 5: Add an HTTPGet node to fetch the content from the URL (`fetcher`) and connect it to the Literal node",
-    "Step 6: Add an LLMAgent node with system prompt 'You are a technical writer. Summarise the text in 3 bullet points.' and model NIM_MODEL (`summarise`), connecting its input to the output of `fetcher`",
-    "Step 7: Add a FileWrite node to write the summary to a file named 'summary.txt' (`writer`), connecting its input to the output of `summarise`",
-    "Step 8: Cook the graph starting from the writer node and print the path where the summary is written"
+    "Step 1: Import necessary modules from `_bootstrap` and `blacknode`: Run `from _bootstrap import NIM_MODEL, require_nim_api_key; import blacknode as bn`",
+    "Step 2: Require NVIDIA NIM API key: Call `require_nim_api_key()` to ensure the API key is set.",
+    "Step 3: Create a graph instance: Initialize `g = bn.Graph()`.",
+    "Step 4: Add nodes for URL, HTTPGet, summarisation, and file writing: `url = g.node('Literal', value='URL of the Wikipedia page'); fetcher = g.node('HTTPGet'); summarise = g.node('LLMAgent', system='You are a technical writer. Summarise the text in 3 bullet points.', model=NIM_MODEL); writer = g.node('FileWrite', path='summary.txt')`",
+    "Step 5: Connect nodes with edges: `url.out('value') >> fetcher.inp('url'); fetcher.out('text') >> summarise.inp('prompt'); summarise.out('text') >> writer.inp('text')`",
+    "Step 6: Cook the graph to execute and get output: `result = g.cook(writer, 'path'); print(f'Summary written to: {result}')`"
   ],
   "outputs": [
-    "Path to the summary file"
+    "Path of the summary file"
   ],
   "failure_modes": [
-    "If the URL is invalid, HTTPGet will fail; if NIM API key is missing, LLMAgent will not function properly"
+    "If the URL is invalid or unreachable, the HTTPGet node will fail; if the summarisation fails, the output text might be empty"
   ],
   "confidence": 0.95,
-  "explanation": "This workflow can be adapted to fetch and summarise any text from a URL using an AI agent and save the summary to a file.",
+  "explanation": "This workflow can be adapted to fetch and summarise any Wikipedia page or similar content source.",
   "source_repo": "https://github.com/temiroff/Blacknode.git",
   "score": 1.0
 }

skills/unifai-workflow-execution/SKILL.md

@@ -1,96 +1,62 @@
 ---
 name: unifai-workflow-execution
 version: 1.0.0
-description: Execute a multi-agent workflow on the UnifAI platform using a specified
-  blueprint and user prompt.
+description: Execute a multi-agent AI workflow defined in YAML or through the UI's
+  drag-and-drop editor.
 inputs:
-- blueprint_id or blueprint_name
-- user_shortcut
-- user_question
+- name: blueprint_path
+  description: Path to the blueprint file (YAML) defining the multi-agent workflow.
+- name: execution_mode
+  description: 'Execution mode: ''local'' or ''distributed''.'
 steps:
-- 'Step 1: Resolve the blueprint ID from either direct ID or name lookup (resolve_blueprint_id
-  method)'
-- 'Step 2: Create a new session from the blueprint (create_session method)'
-- 'Step 3: Submit the session for background execution with the user prompt (submit_session
-  method)'
-- 'Step 4: Poll session status until execution completes (poll_session_status method)'
+- step_name: Load Blueprint
+  description: Parse and validate the blueprint file to ensure it conforms to expected
+    structure.
+- step_name: Initialize Execution Engine
+  description: Set up the execution engine based on the selected mode ('local' or
+    'distributed').
+- step_name: Execute Workflow
+  description: Run the multi-agent workflow, streaming node-by-node output as NDJSON
+    over HTTP.
+- step_name: Stream Results
+  description: Render and stream results in real time to clients subscribing to the
+    event stream.
 outputs:
-- session_id
-- workflow_id
+- name: execution_results
+  description: The output of the executed workflow, streamed as NDJSON over HTTP.
 tags: []
 metadata:
   source_repo: https://github.com/redhat-community-ai-tools/UnifAI.git
   extracted_at: ''
-  confidence: 0.95
+  confidence: 0.9
 ---
 
 # unifai-workflow-execution
 
-Execute a multi-agent workflow on the UnifAI platform using a specified blueprint and user prompt.
-
-## Setup
-
-**Dependencies:**
-
-```text
-pip install requests urllib3
-```
-
-**Setup steps:**
-
-1. Install required dependencies using pip install requests urllib3
-1. Ensure the environment variables are set correctly (BLUEPRINT_ID, BLUEPRINT_NAME, USER_SHORTCUT, POLLING_INTERVAL, UNIFAI_BASE_URL)
-
-## Key Files
-
-- `scripts/execution_workflow.py - Main script for workflow execution`
+Execute a multi-agent AI workflow defined in YAML or through the UI's drag-and-drop editor.
 
 ## Steps
 
-1. Step 1: Resolve the blueprint ID from either direct ID or name lookup (resolve_blueprint_id method)
-2. Step 2: Create a new session from the blueprint (create_session method)
-3. Step 3: Submit the session for background execution with the user prompt (submit_session method)
-4. Step 4: Poll session status until execution completes (poll_session_status method)
-
-## Implementation Details
-
-```python
-resolve_blueprint_id(client: UnifAIClient) -> str
-{...}
-# Resolve the blueprint ID from either direct ID or name lookup.
-```
-
-```python
-create_session(client: UnifAIClient, blueprint_id: str) -> str
-{...}
-# Create a new session from the blueprint.
-```
-
-```python
-submit_session(client: UnifAIClient, session_id: str) -> dict
-{...}
-# Submit the session for background execution with the user prompt.
-```
+1. {'step_name': 'Load Blueprint', 'description': 'Parse and validate the blueprint file to ensure it conforms to expected structure.'}
+2. {'step_name': 'Initialize Execution Engine', 'description': "Set up the execution engine based on the selected mode ('local' or 'distributed')."}
+3. {'step_name': 'Execute Workflow', 'description': 'Run the multi-agent workflow, streaming node-by-node output as NDJSON over HTTP.'}
+4. {'step_name': 'Stream Results', 'description': 'Render and stream results in real time to clients subscribing to the event stream.'}
 
 ## Inputs
 
-- blueprint_id or blueprint_name
-- user_shortcut
-- user_question
+- {'name': 'blueprint_path', 'description': 'Path to the blueprint file (YAML) defining the multi-agent workflow.'}
+- {'name': 'execution_mode', 'description': "Execution mode: 'local' or 'distributed'."}
 
 ## Outputs
 
-- session_id
-- workflow_id
+- {'name': 'execution_results', 'description': 'The output of the executed workflow, streamed as NDJSON over HTTP.'}
 
 ## Failure Modes
 
-- Blueprint name not found or not unique - error during blueprint resolution
-- Session creation fails - error from API response
-- Session submission fails - error from API response
-- Polling session status fails - error from API response
+- {'mode_name': 'Invalid Blueprint', 'description': 'Blueprint file is not valid YAML or does not conform to expected structure.'}
+- {'mode_name': 'Execution Engine Initialization Failure', 'description': 'Failed to initialize the execution engine due to configuration issues or missing dependencies.'}
 
 ## Source
 
 Extracted from: [https://github.com/redhat-community-ai-tools/UnifAI.git](https://github.com/redhat-community-ai-tools/UnifAI.git)
-Confidence: 0.95
+Confidence: 0.9

skills/unifai-workflow-execution/examples.md

@@ -4,7 +4,7 @@
 
 ```python
 # How to use this skill
-# Inputs: blueprint_id or blueprint_name, user_shortcut, user_question
-# Process: Step 1: Resolve the blueprint ID from either direct ID or name lookup (resolve_blueprint_id method) → Step 2: Create a new session from the blueprint (create_session method) → Step 3: Submit the session for background execution with the user prompt (submit_session method)
-# Outputs: session_id, workflow_id
+# Inputs: {'name': 'blueprint_path', 'description': 'Path to the blueprint file (YAML) defining the multi-agent workflow.'}, {'name': 'execution_mode', 'description': "Execution mode: 'local' or 'distributed'."}
+# Process: {'step_name': 'Load Blueprint', 'description': 'Parse and validate the blueprint file to ensure it conforms to expected structure.'} → {'step_name': 'Initialize Execution Engine', 'description': "Set up the execution engine based on the selected mode ('local' or 'distributed')."} → {'step_name': 'Execute Workflow', 'description': 'Run the multi-agent workflow, streaming node-by-node output as NDJSON over HTTP.'}
+# Outputs: {'name': 'execution_results', 'description': 'The output of the executed workflow, streamed as NDJSON over HTTP.'}
 ```

skills/unifai-workflow-execution/metadata.json

@@ -1,30 +1,53 @@
 {
   "name": "unifai-workflow-execution",
   "version": "1.0.0",
-  "goal": "Execute a multi-agent workflow on the UnifAI platform using a specified blueprint and user prompt.",
+  "goal": "Execute a multi-agent AI workflow defined in YAML or through the UI's drag-and-drop editor.",
   "inputs": [
-    "blueprint_id or blueprint_name",
-    "user_shortcut",
-    "user_question"
+    {
+      "name": "blueprint_path",
+      "description": "Path to the blueprint file (YAML) defining the multi-agent workflow."
+    },
+    {
+      "name": "execution_mode",
+      "description": "Execution mode: 'local' or 'distributed'."
+    }
   ],
   "steps": [
-    "Step 1: Resolve the blueprint ID from either direct ID or name lookup (resolve_blueprint_id method)",
-    "Step 2: Create a new session from the blueprint (create_session method)",
-    "Step 3: Submit the session for background execution with the user prompt (submit_session method)",
-    "Step 4: Poll session status until execution completes (poll_session_status method)"
+    {
+      "step_name": "Load Blueprint",
+      "description": "Parse and validate the blueprint file to ensure it conforms to expected structure."
+    },
+    {
+      "step_name": "Initialize Execution Engine",
+      "description": "Set up the execution engine based on the selected mode ('local' or 'distributed')."
+    },
+    {
+      "step_name": "Execute Workflow",
+      "description": "Run the multi-agent workflow, streaming node-by-node output as NDJSON over HTTP."
+    },
+    {
+      "step_name": "Stream Results",
+      "description": "Render and stream results in real time to clients subscribing to the event stream."
+    }
   ],
   "outputs": [
-    "session_id",
-    "workflow_id"
+    {
+      "name": "execution_results",
+      "description": "The output of the executed workflow, streamed as NDJSON over HTTP."
+    }
   ],
   "failure_modes": [
-    "Blueprint name not found or not unique - error during blueprint resolution",
-    "Session creation fails - error from API response",
-    "Session submission fails - error from API response",
-    "Polling session status fails - error from API response"
+    {
+      "mode_name": "Invalid Blueprint",
+      "description": "Blueprint file is not valid YAML or does not conform to expected structure."
+    },
+    {
+      "mode_name": "Execution Engine Initialization Failure",
+      "description": "Failed to initialize the execution engine due to configuration issues or missing dependencies."
+    }
   ],
-  "confidence": 0.95,
-  "explanation": "This workflow is specific to the UnifAI platform and its multi-agent system, but can be adapted for similar systems with a similar architecture.",
+  "confidence": 0.9,
+  "explanation": "This workflow is designed to execute multi-agent AI workflows defined in YAML blueprints or through the UI's drag-and-drop editor, providing real-time streaming of results.",
   "source_repo": "https://github.com/redhat-community-ai-tools/UnifAI.git",
   "score": 1.0
 }