github/ChatDev
1
0
Fork 0
mirror of https://github.com/OpenBMB/ChatDev.git synced 2026-06-01 05:00:40 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update GameDev_v1.yaml

Browse Source 
refined prompts, removed QA counter
This commit is contained in:
kilo2127 2026-01-13 23:23:29 +08:00
parent 2e65ba7c66
commit e96ed42602
1 changed files with 173 additions and 142 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

191
yaml_instance/GameDev_v1.yaml
View File

@ -1,9 +1,8 @@
graph: graph:
id: pygame_game_factory_streamlined id: pygame_game_factory_streamlined
description: "Streamlined two-phase workflow to generate cool, playable Pygame games with visual polish." description: Streamlined two-phase workflow to generate cool, playable Pygame games with visual polish.
log_level: INFO log_level: INFO
is_majority_voting: false is_majority_voting: false
nodes: nodes:
- id: Game Designer - id: Game Designer
type: agent type: agent
@ -66,7 +65,6 @@ graph:
# Constraints # Constraints
- DO NOT suggest loading external assets (images, fonts, sounds). - DO NOT suggest loading external assets (images, fonts, sounds).
- Keep the scope small enough for a 200-line single-file implementation. - Keep the scope small enough for a 200-line single-file implementation.
- id: Planner - id: Planner
type: agent type: agent
context_window: 0 context_window: 0
@ -129,10 +127,9 @@ graph:
4. Animation & Juice: Add smooth movements, size pulsing, rotation. 4. Animation & Juice: Add smooth movements, size pulsing, rotation.
5. Visual Feedback: Flash effects on collisions, spawn animations. 5. Visual Feedback: Flash effects on collisions, spawn animations.
6. UI Polish: Centered text, drop shadows, background gradients. 6. UI Polish: Centered text, drop shadows, background gradients.
- id: Core_Developer - id: Core_Developer
type: agent type: agent
context_window: 5 context_window: 6
config: config:
provider: openai provider: openai
name: gpt-4o name: gpt-4o
@ -182,11 +179,25 @@ graph:
- NO visual effects (particles, shake, etc.) in Phase 1 - NO visual effects (particles, shake, etc.) in Phase 1
PROCESS: PROCESS:
1. Write complete, runnable code 1. Write the complete, runnable game code based on the plan
2. Call `save_file` with path="game.py" and full code as content
2. Call `save_file` with TWO REQUIRED ARGUMENTS:
save_file(
path="game.py",
content="<COMPLETE PYTHON CODE AS STRING>"
)
CRITICAL: The 'content' parameter MUST contain the full game code.
DO NOT call save_file with empty arguments.
3. Output exactly: "Phase 1 complete. Code saved to game.py." 3. Output exactly: "Phase 1 complete. Code saved to game.py."
DO NOT output code in your message. DO NOT add explanations. IMPORTANT OUTPUT RULES:
- You MUST generate the complete Python code internally
- Pass this code as a STRING to save_file's 'content' parameter in the tool call
- DO NOT paste the code in your chat message text (to save tokens and avoid duplication)
- The code should ONLY appear inside the tool call's 'content' argument, NOT in your message
- DO NOT add explanations after calling save_file
tooling: tooling:
- type: function - type: function
config: config:
@ -195,10 +206,9 @@ graph:
- name: install_python_packages - name: install_python_packages
- name: save_file - name: save_file
- name: describe_available_files - name: describe_available_files
- id: Polish_Developer - id: Polish_Developer
type: agent type: agent
context_window: 3 context_window: 6
config: config:
provider: openai provider: openai
name: gpt-4o name: gpt-4o
@ -314,14 +324,39 @@ graph:
4. pygame.QUIT handled properly 4. pygame.QUIT handled properly
PROCESS: PROCESS:
1. Call `read_text_file_snippet` to read game.py (use limit=25000, check truncation) 1. Call `read_text_file_snippet(path="game.py", offset=0, limit=25000)` to read the code
2. Apply ALL visual design principles above - If truncated=true in response, call again with offset=25000 to get the rest
3. Preserve ALL game logic - DO NOT break gameplay - Store the COMPLETE code string in memory
4. Verify all 4 mandatory requirements are present
5. Call `save_file` with path="game.py" and complete enhanced code
6. Output exactly: "Phase 2 complete. Code saved to game.py."
DO NOT output code in your message. DO NOT add descriptions. 2. In your reasoning, plan the visual enhancements:
- Identify which color palette to use based on game theme
- Decide which particle effects to add (subtle, not overwhelming)
- Plan animation improvements (smooth, not bouncy)
3. Construct the complete enhanced code as a single Python string:
- Apply the color palette throughout
- Add particle system code
- Add screen shake and animation code
- Preserve ALL game logic (do not break functionality)
4. Call `save_file` with TWO REQUIRED ARGUMENTS:
save_file(
path="game.py",
content="<THE COMPLETE PYTHON CODE AS A STRING>"
)
CRITICAL: The 'content' parameter MUST be the full Python code string.
DO NOT call save_file with empty arguments.
DO NOT call save_file multiple times.
5. After successful save, output exactly: "Phase 2 complete. Code saved to game.py."
IMPORTANT OUTPUT RULES:
- You MUST generate the complete enhanced Python code internally
- Pass this code as a STRING to save_file's 'content' parameter in the tool call
- DO NOT paste the code in your chat message text (to save tokens and avoid duplication)
- The code should ONLY appear inside the tool call's 'content' argument, NOT in your message
- DO NOT add explanations after calling save_file
tooling: tooling:
- type: function - type: function
config: config:
@ -330,7 +365,6 @@ graph:
- name: read_text_file_snippet - name: read_text_file_snippet
- name: save_file - name: save_file
- name: describe_available_files - name: describe_available_files
- id: Game_Launcher - id: Game_Launcher
type: agent type: agent
context_window: 0 context_window: 0
@ -370,16 +404,14 @@ graph:
auto_load: true auto_load: true
tools: tools:
- name: read_text_file_snippet - name: read_text_file_snippet
- id: Final_Game_Executor - id: Final_Game_Executor
type: python type: python
config: config:
timeout_seconds: 120 timeout_seconds: 120
encoding: utf-8 encoding: utf-8
- id: Bug_Fixer - id: Bug_Fixer
type: agent type: agent
context_window: 5 context_window: 6
config: config:
provider: openai provider: openai
name: gpt-4o name: gpt-4o
@ -397,16 +429,33 @@ graph:
PROCESS: PROCESS:
1. Read the error traceback from input (identifies line number and error type) 1. Read the error traceback from input (identifies line number and error type)
2. Read the COMPLETE game.py (make multiple read calls if needed)
2. Read the COMPLETE game.py file:
- Call `read_text_file_snippet(path="game.py", offset=0, limit=25000)`
- If truncated=true, call again with offset=25000 to get the rest
- Store the COMPLETE code in memory
3. Verify file completeness: 3. Verify file completeness:
- Check that code doesn't end mid-line or mid-function - Check that code doesn't end mid-line or mid-function
- Verify main() function exists - Verify main() function exists
- Verify if __name__ == "__main__" block exists - Verify if __name__ == "__main__" block exists
- If incomplete, read more until complete - If incomplete, read more until complete
4. Identify the EXACT line(s) causing the error
5. Fix ONLY those specific line(s) using minimal changes 4. Identify the EXACT line(s) causing the error and fix them:
6. Save the COMPLETE fixed code (ALL original lines + the fix) - Make ONLY minimal changes to fix the specific error
7. Output: "Bug fixed: [one-line description]. Code saved to game.py." - Keep all other code EXACTLY as is
5. Call `save_file` with TWO REQUIRED ARGUMENTS:
save_file(
path="game.py",
content="<COMPLETE FIXED CODE AS STRING>"
)
CRITICAL: The 'content' parameter MUST be the complete fixed code.
DO NOT call save_file with empty arguments.
DO NOT save incomplete code.
6. Output: "Bug fixed: [one-line description]. Code saved to game.py."
COMMON FIXES (minimal changes only): COMMON FIXES (minimal changes only):
- IndentationError: Fix spacing on the error line only - IndentationError: Fix spacing on the error line only
@ -421,7 +470,12 @@ graph:
- DO NOT simplify or optimize code - DO NOT simplify or optimize code
- DO NOT save if file is incomplete (< 10000 bytes is suspicious) - DO NOT save if file is incomplete (< 10000 bytes is suspicious)
OUTPUT: Only a brief fix description, NO code blocks in your message. IMPORTANT OUTPUT RULES:
- You MUST generate the complete fixed Python code internally
- Pass this code as a STRING to save_file's 'content' parameter in the tool call
- DO NOT paste the code in your chat message text (to save tokens and avoid duplication)
- The code should ONLY appear inside the tool call's 'content' argument, NOT in your message
- Only output a brief fix description in your message, NO code blocks
tooling: tooling:
- type: function - type: function
config: config:
@ -429,10 +483,9 @@ graph:
tools: tools:
- name: read_text_file_snippet - name: read_text_file_snippet
- name: save_file - name: save_file
- id: QA_Agent - id: QA_Agent
type: agent type: agent
context_window: 5 context_window: 6
config: config:
provider: openai provider: openai
name: gpt-4o name: gpt-4o
@ -504,10 +557,9 @@ graph:
auto_load: true auto_load: true
tools: tools:
- name: read_text_file_snippet - name: read_text_file_snippet
- id: Polish_Refinement - id: Polish_Refinement
type: agent type: agent
context_window: 5 context_window: 6
config: config:
provider: openai provider: openai
name: gpt-4o name: gpt-4o
@ -519,15 +571,35 @@ graph:
PROCESS: PROCESS:
1. Read QA feedback (lists specific issues) 1. Read QA feedback (lists specific issues)
2. Read COMPLETE game.py (use limit=25000, check truncation)
2. Read the COMPLETE game.py file:
- Call `read_text_file_snippet(path="game.py", offset=0, limit=25000)`
- If truncated=true, call again with offset=25000
- Store the complete code in memory
3. Fix ONLY the identified issues: 3. Fix ONLY the identified issues:
- If color issue: adjust palette - If color issue: adjust palette
- If animation issue: add/improve effects - If animation issue: add/improve effects
- If UI issue: redesign layout/text - If UI issue: redesign layout/text
- If functional issue: add missing feature - If functional issue: add missing feature
4. Verify ALL other code remains unchanged - Verify ALL other code remains unchanged
5. Save complete fixed code
6. Output: "Refinement complete. Issues addressed: [list]. Code saved." 4. Call `save_file` with TWO REQUIRED ARGUMENTS:
save_file(
path="game.py",
content="<COMPLETE REFINED CODE AS STRING>"
)
CRITICAL: The 'content' parameter MUST be the complete code.
DO NOT call save_file with empty arguments.
5. Output: "Refinement complete. Issues addressed: [list]. Code saved."
IMPORTANT OUTPUT RULES:
- You MUST generate the complete refined Python code internally
- Pass this code as a STRING to save_file's 'content' parameter in the tool call
- DO NOT paste the code in your chat message text (to save tokens and avoid duplication)
- The code should ONLY appear inside the tool call's 'content' argument, NOT in your message
CONSTRAINTS: CONSTRAINTS:
- Make MINIMAL changes (only fix what QA mentioned) - Make MINIMAL changes (only fix what QA mentioned)
@ -541,52 +613,32 @@ graph:
tools: tools:
- name: read_text_file_snippet - name: read_text_file_snippet
- name: save_file - name: save_file
- id: QA_Loop_Counter
type: loop_counter
config:
max_iterations: 3
reset_on_emit: false
message: "Maximum QA refinement attempts (3) reached. Proceeding with current version."
- id: Bug_Fix_Counter
type: loop_counter
config:
max_iterations: 3
reset_on_emit: false
message: "Maximum bug fix attempts (3) reached. Game has critical bugs that could not be fixed automatically."
edges: edges:
- from: Game Designer - from: Game Designer
to: Planner to: Planner
trigger: true trigger: true
carry_data: true carry_data: true
- from: Planner - from: Planner
to: Core_Developer to: Core_Developer
trigger: true trigger: true
carry_data: true carry_data: true
keep_message: true keep_message: true
- from: Core_Developer - from: Core_Developer
to: Polish_Developer to: Polish_Developer
trigger: true trigger: true
carry_data: true carry_data: true
- from: Polish_Developer - from: Polish_Developer
to: QA_Agent to: QA_Agent
trigger: true trigger: true
carry_data: true carry_data: true
- from: QA_Agent - from: QA_Agent
to: QA_Loop_Counter to: Polish_Refinement
condition: condition:
type: keyword type: keyword
config: config:
any: any:
- NEEDS_REFINEMENT - NEEDS_REFINEMENT
carry_data: true carry_data: true
- from: QA_Agent - from: QA_Agent
to: Game_Launcher to: Game_Launcher
condition: condition:
@ -595,47 +647,26 @@ graph:
any: any:
- APPROVE - APPROVE
carry_data: false carry_data: false
- from: QA_Loop_Counter
to: Polish_Refinement
condition: loop_pass
carry_data: true
- from: Polish_Refinement - from: Polish_Refinement
to: QA_Agent to: Game_Launcher
trigger: true trigger: true
carry_data: false carry_data: false
- from: QA_Loop_Counter
to: Game_Launcher
condition: loop_fail
carry_data: false
- from: Game_Launcher - from: Game_Launcher
to: Final_Game_Executor to: Final_Game_Executor
trigger: true trigger: true
carry_data: true carry_data: true
- from: Final_Game_Executor - from: Final_Game_Executor
to: Bug_Fix_Counter to: Bug_Fixer
condition: code_fail condition: code_fail
carry_data: true carry_data: true
- from: Bug_Fix_Counter
to: Bug_Fixer
condition: loop_pass
carry_data: true
- from: Bug_Fixer - from: Bug_Fixer
to: Game_Launcher to: Game_Launcher
trigger: true trigger: true
carry_data: false carry_data: false
initial_instruction: Describe a game idea you want to see come to life (e.g., 'A space shooter where you dodge asteroids', 'A rhythm game with falling notes', 'Snake but with gravity'). The system will automatically design, plan, implement core mechanics, add visual polish, perform QA checks, and execute the game.
initial_instruction: "Describe a game idea you want to see come to life (e.g., 'A space shooter where you dodge asteroids', 'A rhythm game with falling notes', 'Snake but with gravity'). The system will automatically design, plan, implement core mechanics, add visual polish, perform QA checks, and execute the game."
start: start:
- Game Designer - Game Designer
end: end:
- Final_Game_Executor - Final_Game_Executor
version: 1.0.0 version: 1.0.0
vars: {} vars: {}