3b250c95df
Quick test mode for simulation tests Fixed o4-mini name, OpenAI removed o4-mini-high Add max_output_tokens property to ModelCapabilities Fixed tests after refactor
Fahad
2025-06-23 18:56:47 +04:00
ce6c1fd7ea
Quick test mode for simulation tests Fixed o4-mini name, OpenAI removed o4-mini-high Add max_output_tokens property to ModelCapabilities
Fahad
2025-06-23 18:33:47 +04:00
8c1814d4eb
Quick test mode for simulation tests
Fahad
2025-06-23 18:05:31 +04:00
9167e6d845
Quick test mode for simulation tests
Fahad
2025-06-23 17:53:03 +04:00
14eaf930ed
Cleanup, use ModelCapabilities only
Fahad
2025-06-23 17:39:47 +04:00
498ea88293
Use ModelCapabilities consistently instead of dictionaries Moved aliases as part of SUPPORTED_MODELS instead of shorthand, more in line with how custom_models are declared Further refactoring to cleanup some code
Fahad
2025-06-23 16:58:59 +04:00
521c6c0e61
Improved consensus to treat a step properly as both a request + response, and initial step includes Claude's assessment. Improved prompt to not request for code when it's a general business decision
Fahad
2025-06-22 13:37:32 +04:00
18f6f16ac6
Improved consensus to treat a step properly as both a request + response, and initial step includes Claude's assessment. Improved prompt to not request for code when it's a general business decision
Fahad
2025-06-22 13:21:09 +04:00
355331d141
Exclude 'model' parameter for consensus as it uses its own
Fahad
2025-06-22 12:22:04 +04:00
14d1923de8
During documentation must stop on discovery of critical logic errors in code and inform user
Fahad
2025-06-22 10:51:28 +04:00
e8275a04a0
Fix model metadata preservation with continuation_id (fixes#111) (#113)
Brad Fair
2025-06-22 01:28:58 -05:00
81464ec6c6
Updated with final flow
Fahad
2025-06-22 10:27:37 +04:00
fccfb0d999
Re-imagined and re-written Debug tool. Instead of prompting Claude to perform initial analysis (and hoping it did), the tool now works through the debug process as an 'investigation', encouraging Claud to gather its 'findings' / 'hypothesis', stepping back as needed, collecting files it's gone through and keeping track of files relevant to the issue at hand. This structured investiion is then passed to the other model with far greater insight than the original debug tool ever could. Improved prompts, guard against overengineering and flag that as an antipattern
Fahad
2025-06-19 10:22:30 +04:00
e74e612de9
Keep last
Fahad
2025-06-18 05:43:29 +04:00
5c75ac978d
AI banter
Fahad
2025-06-18 05:42:55 +04:00
ee52d7da4e
Fixed links for Planner tool
Fahad
2025-06-17 20:52:35 +04:00
a509730dca
New Planner tool to help you break down complex ideas, problems, and projects into multiple manageable steps. This is a self-prompt generation tool whose output can then be fed into another tool and model as required
Fahad
2025-06-17 20:49:53 +04:00
3667ed3a43
Steer the LLMs away from being overtly assertive in trying to discover non-existent bugs during a debug investigation session Add a no_bug_found response
Fahad
2025-06-17 17:01:46 +04:00
044a8621a3
Vastly improved debug tool and related instructions Accompanying simulation test Cleanup - A single source of truth for parameter descriptions
Fahad
2025-06-17 16:23:26 +04:00
9bf2a2a51c
Improved prompt in case Claude tries to use a smaller model with a larger file
Fahad
2025-06-17 13:08:50 +04:00
b528598360
Add regression tests for Gemini parameter order bug
Fahad
2025-06-16 19:13:47 +04:00
f55f2b0a0f
Fix Google model restriction parameter order regression (#62)
Ming
2025-06-16 23:12:16 +08:00
70b64adff3
Schema now lists all models including locally available models New tool to list all models listmodels Integration test to for all the different combinations of API keys Tweaks to codereview prompt for a better quality input from Claude Fixed missing 'low' severity in codereview
Fahad
2025-06-16 19:07:35 +04:00
OhMyApps