feat: enhance agent message with stream_writer

[vcfgv]

No description provided.

[Copilot]

Pull Request Overview

Enhances agent message functionality with stream_writer to provide real-time status updates during financial analysis operations. This adds transparency to the hedge fund analysis process by streaming progress information to users as agents perform their financial calculations.

• Adds get_stream_writer() usage to all financial agents for progress streaming
• Updates the example client to handle streaming response format with task and event data
• Upgrades langgraph dependency to support enhanced streaming capabilities

Reviewed Changes

Copilot reviewed 21 out of 23 changed files in this pull request and generated 1 comment.

File	Description
python/valuecell/examples/core_remote_agent_demo.py	Updates client to handle streaming tuple format (task, event)
python/third_party/ai-hedge-fund/src/agents/*.py	Adds stream writer calls throughout all 15+ agent files for progress updates
python/third_party/ai-hedge-fund/pyproject.toml	Updates langgraph to v0.3.34 and removes a2a-sdk dependency
python/third_party/ai-hedge-fund/adapter/main.py	Implements streaming workflow with proper error handling

Comments suppressed due to low confidence (4)

python/third_party/ai-hedge-fund/src/agents/stanley_druckenmiller.py:1

• Multiple identical writer messages 'Analyzing for {ticker}...' make it impossible to distinguish which specific analysis step is being performed. Each should describe the specific analysis type (sentiment, insider activity, risk-reward, etc.).

from src.graph.state import AgentState, show_agent_reasoning

python/third_party/ai-hedge-fund/src/agents/rakesh_jhunjhunwala.py:1

• Similar issue with generic writer messages. Each analysis step (growth, profitability, balance sheet, cash flow, etc.) should have descriptive messages that match the progress.update_status calls.

from src.graph.state import AgentState, show_agent_reasoning

python/third_party/ai-hedge-fund/src/agents/cathie_wood.py:1

• [nitpick] Generic messages like 'Analyzing for {ticker}...' and 'Calculating for {ticker}...' don't provide meaningful progress information. Consider using specific descriptions like 'Analyzing disruptive potential for {ticker}...' or 'Calculating valuation & high-growth scenario for {ticker}...'.

from src.graph.state import AgentState, show_agent_reasoning

python/third_party/ai-hedge-fund/src/agents/bill_ackman.py:1

• [nitpick] The generic message 'Generating for {ticker}...' should be more specific, like 'Generating Bill Ackman analysis for {ticker}...' to match the progress update and provide clear feedback to users.

from src.graph.state import AgentState, show_agent_reasoning

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}