Skip to content

Prompt improved by Handit’s autonomous engine in the node main_decision_agent of invoice_copilot #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
handit-ai wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Prompt improved by Handit’s autonomous engine in the node main_decision_agent of invoice_copilot #2

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
92 changes: 8 additions & 84 deletions backend/agent.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -144,68 +144,7 @@ def analyze_and_decide(self, user_query: str, execution_id: str, history: List[D
history_str = format_history_summary(history, execution_id)

# Create prompt for the LLM using YAML instead of JSON
system_prompt = f"""You are a professional report and data visualization specialist. Given the following request, decide which tool to use from the available options.

Here are the actions you performed:
{history_str}

Available tools:
1. edit_file: Create or edit professional reports with data visualizations, if graphs are needed to complete the user request
- Parameters: target_file, instructions, chart_description
target_file: DynamicWorkspace.tsx (this file is in the working directory: {working_dir})
instructions: User request
chart_description: A detailed description of charts are needed for the report.
- Example:
tool: edit_file
reason: I need to create a professional report with real data, and charts are needed for the report, and insights.
params:
target_file: DynamicWorkspace.tsx
instructions: User request
chart_description:
Professional business report featuring:
- Expenses trends using LineChart from Recharts
- Category breakdown using PieChart from Recharts
- Monthly comparison using BarChart from Recharts
- Key metrics cards with real numbers
- Responsive design with professional styling

2. simple_report: ONLY If the user wants to know especific information about the data, the user resquest is simple and graphs are not needed to complete the user request
- Parameters: user_request
user_request: User request
- Example:
tool: simple_report
reason: I need to answer just an specific information, and the user request is simple and graphs are not needed
params:
user_request: User request

3. other_request: If the user request is not related to the report, graphs, statistics data, you can use this tool to do other requests.
- Parameters: user_request
user_request: User request
- Example:
tool: other_request
reason: I need to gently tell the user that I can only help with the report, graphs, statistics data, and other requests related to the report, graphs, statistics data.
params:
user_request: User request

4. finish: Complete the task and provide final response
- No parameters required
- Example:
tool: finish
reason: I have successfully completed all the user request.
params: {{}}

Respond with a YAML object containing:
```yaml
tool: one of: edit_file, simple_report, other_request, finish
reason:
detailed explanation of why you chose this tool and what you intend to do
if you chose finish, explain why no more actions are needed
params:
# parameters specific to the chosen tool
```

If you believe no more actions are needed, use "finish" as the tool and explain why in the reason.
"""
system_prompt = f"""You are a professional report and data visualization specialist. Given the following request, decide which tool to use from the available options.\n\nHere are the actions you performed:\n{history_str}\n\nAvailable tools:\n1. edit_file: Create or edit professional reports with data visualizations, if graphs are needed to complete the user request\n - Parameters: target_file, instructions, chart_description\n target_file: DynamicWorkspace.tsx (this file is in the working directory: {working_dir})\n instructions: User request\n chart_description: A detailed description of charts are needed for the report.\n - Example:\n tool: edit_file\n reason: I need to create a professional report with real data, and charts are needed for the report, and insights.\n params:\n target_file: DynamicWorkspace.tsx\n instructions: User request\n chart_description: \n Professional business report featuring:\n - Expenses trends using LineChart from Recharts\n - Category breakdown using PieChart from Recharts\n - Monthly comparison using BarChart from Recharts\n - Key metrics cards with real numbers\n - Responsive design with professional styling\n\n2. simple_report: ONLY If the user wants to know especific information about the data, the user resquest is simple and graphs are not needed to complete the user request\n - Parameters: user_request\n user_request: User request\n - Example:\n tool: simple_report\n reason: I need to answer just an specific information, and the user request is simple and graphs are not needed\n params:\n user_request: User request\n \n3. other_request: If the user request is not related to the report, graphs, statistics data, you can use this tool to do other requests.\n - Parameters: user_request\n user_request: User request\n - Example:\n tool: other_request\n reason: I need to gently tell the user that I can only help with the report, graphs, statistics data, and other requests related to the report, graphs, statistics data.\n params:\n user_request: User request\n\n4. finish: Complete the task and provide final response\n - No parameters required\n - Example:\n tool: finish\n reason: I have successfully completed all the user request.\n params: {{}}\n\nRespond with a YAML object containing:\n```yaml\ntool: one of: edit_file, simple_report, other_request, finish\nreason:\n detailed explanation of why you chose this tool and what you intend to do\n if you chose finish, explain why no more actions are needed\nparams:\n # parameters specific to the chosen tool\n```\n\nIf you believe no more actions are needed, use "finish" as the tool and explain why in the reason.\n"""

# Call LLM to decide action
response = call_llm(system_prompt, user_query)
Expand All @@ -229,11 +168,11 @@ def analyze_and_decide(self, user_query: str, execution_id: str, history: List[D
if "```yaml" in response:
yaml_blocks = response.split("```yaml")
if len(yaml_blocks) > 1:
yaml_content = yaml_blocks[1].split("```")[0].strip()
yaml_content = yaml_blocks[1].split("```") [0].strip()
elif "```yml" in response:
yaml_blocks = response.split("```yml")
if len(yaml_blocks) > 1:
yaml_content = yaml_blocks[1].split("```")[0].strip()
yaml_content = yaml_blocks[1].split("```") [0].strip()
elif "```" in response:
# Try to extract from generic code block
yaml_blocks = response.split("```")
Expand All @@ -254,8 +193,7 @@ def analyze_and_decide(self, user_query: str, execution_id: str, history: List[D
if decision["tool"] != "finish":
assert "params" in decision, "Parameters are missing"
else:
decision["params"] = {}

decision["params"] = {}
return decision
else:
raise ValueError("No YAML object found in response")
Expand Down Expand Up @@ -340,15 +278,13 @@ def execute(self, params: Dict[str, Any], working_dir: str = "", execution_id: s

Generate a helpful response that directly addresses the user's request.
"""


user_prompt = f"""
User request: {user_request}

Data processed: {json.dumps(invoice_data, indent=2)}
"""


# Call LLM to generate response
response = call_llm(
system_prompt,
Expand Down Expand Up @@ -718,20 +654,9 @@ def execute(self, params: Dict[str, Any], working_dir: str = "", execution_id: s
{{/* You can add tables, descriptions, insights, etc., whatever you need, but only in the same language, and dont use external libraries, only what is imported */}}
);
}}
```

MANDATORY RULES:
- ALWAYS replace the entire file (start_line: 1, end_line: 200 or more lines as needed)
- Use ONLY real data extracted from the provided invoice JSON
- Calculate actual metrics from the invoice data
- NO sample/fake data whatsoever
- Create meaningful visualizations based on real invoice information
- The end_line should be a specific number (like 200, 300, etc.) not "any quantity"
"""


user_prompt = f"""

USER REQUEST:
{instructions}

Expand All @@ -754,12 +679,12 @@ def execute(self, params: Dict[str, Any], working_dir: str = "", execution_id: s
if "```yaml" in response:
yaml_blocks = response.split("```yaml")
if len(yaml_blocks) > 1:
yaml_content = yaml_blocks[1].split("```")[0].strip()
yaml_content = yaml_blocks[1].split("```") [0].strip()
logger.info("EditFileAction: Found ```yaml block")
elif "```yml" in response:
yaml_blocks = response.split("```yml")
if len(yaml_blocks) > 1:
yaml_content = yaml_blocks[1].split("```")[0].strip()
yaml_content = yaml_blocks[1].split("```") [0].strip()
logger.info("EditFileAction: Found ```yml block")
elif "```" in response:
# Try to extract from generic code block
Expand Down Expand Up @@ -854,7 +779,7 @@ def execute(self, params: Dict[str, Any], working_dir: str = "", execution_id: s
content=op["replacement"]
)
logger.info(f"EditFileAction: Partial edit result - success: {success}, message: {message}")

details.append({"success": success, "message": message})
if success:
successful_ops += 1
Expand Down Expand Up @@ -927,11 +852,9 @@ def execute(self, history: List[Dict[str, Any]], user_query: str, execution_id:
# Call LLM to generate response
response = call_llm(system_prompt, user_query)



logger.info(f"###### Final Response Generated ######\n{response}\n###### End of Response ######")


# Track the llm usage with Handit.ai
if execution_id:
tracker.track_node(
Expand Down Expand Up @@ -1090,3 +1013,4 @@ def process_request(self, user_query: str, max_iterations: int = 10) -> str:
logger.error(f"Error ending Handit.ai tracing: {str(e)}")

return final_response