0ae6bea9d5
FIX Redundant code block in run_agent_with_retry
2025-03-11 22:49:18 +00:00
9f3e6b9ae8
FIX removed --disable-thinking flag because we have the skip sonnet flag
2025-03-11 22:49:18 +00:00
36678969a9
FEAT automatically detect the sonnet3.7 error and auto apply a workaround
2025-03-11 22:49:18 +00:00
108244d091
FIX Claude 3.7 Sonnet throws Expected thinking
2025-03-11 22:49:16 +00:00
376fe18b83
activity panel
2025-03-11 14:55:43 -04:00
89ee1d96ef
vsc icon
2025-03-11 14:15:37 -04:00
750c0d893b
vscode extension
2025-03-11 13:32:46 -04:00
37764c7d56
fix tests
2025-03-11 11:25:27 -04:00
ae9cf5021b
record trajectory at all steps
2025-03-11 11:01:05 -04:00
5d899d3d13
trajectory
2025-03-11 08:56:12 -04:00
d0985b6b84
store trajectory for tasks
2025-03-11 08:41:59 -04:00
7038a61693
mock repos for test
2025-03-11 08:41:51 -04:00
96093e8dfc
trajectory for stage transitions
2025-03-11 08:34:08 -04:00
b4b0fdd686
Merge pull request #120 from ariel-frischer/cost-display
...
Add Cost Display for Default ReAct Agent
2025-03-10 20:12:01 -04:00
4aa1b0ba93
refactor(agent_utils.py): remove unused run_research_agent and run_web_research_agent functions to clean up code and improve maintainability
...
refactor(agent_utils.py): import run_research_agent and run_web_research_agent from their respective modules to streamline the code structure and enhance clarity
2025-03-10 17:09:05 -07:00
c4b2d2bb51
fix(agent_utils.py): remove merge conflict markers and clean up imports to ensure code integrity
...
refactor(agent_utils.py): consolidate research agent imports for better organization and readability
2025-03-10 17:07:24 -07:00
89e4556e7b
feat: implement trajectory model and repository for tracking agent actions
...
This commit introduces a new `Trajectory` model to the database, which tracks the sequence of actions taken by agents, including tool executions and their results. The addition of the `TrajectoryRepository` allows for storing and retrieving these trajectories, enabling better analysis of agent behavior and debugging of issues.
Additionally, the commit refactors existing code to utilize the new repository and model, improving the overall architecture and maintainability of the codebase. This change is essential for enhancing the capabilities of the agent system and providing a more robust framework for future development.
2025-03-10 17:05:34 -07:00
4d4eb6cadb
fix(tests): add cost_cb parameter to fake_print_agent_output function to enhance flexibility in testing
...
style(tests): add newline at the end of test_handle_api_error_resource_exhausted function for consistency with PEP 8 guidelines
2025-03-10 17:03:42 -07:00
8f2adc7f61
fix(output.py): update subtitle text from "Total Cost" to "Cost" for clarity in cost information display
2025-03-10 17:01:00 -07:00
f43c5e72b6
feat(output.py): add get_cost_subtitle function to generate cost subtitles for agent output
...
refactor(output.py): simplify subtitle generation in print_agent_output function by using get_cost_subtitle
2025-03-10 16:58:39 -07:00
416689b030
feat(agent_utils.py): pass callback to print_agent_output to track costs and tokens
...
feat(output.py): enhance print_agent_output to display cost and token information in the output panel
2025-03-10 16:51:42 -07:00
909825bf1b
refactor: extract get_most_recent_id
2025-03-10 19:37:39 -04:00
b3010bb649
add error info to trajectory records
2025-03-10 19:12:14 -04:00
78983ec20b
add trajectory table
2025-03-10 18:59:42 -04:00
a18998be0d
add project info to reasoning assist prompts
2025-03-10 18:01:56 -04:00
3db7cc2ca9
improve prompt
2025-03-10 17:47:31 -04:00
b5e4c64042
improve prompts; make list_directory more resilient
2025-03-10 16:41:09 -04:00
7d579f5557
improve prompts
2025-03-10 15:37:38 -04:00
e81421a95a
improve prompts
2025-03-10 15:30:22 -04:00
c98c107ce3
use expert model for reasoning assisted mode
2025-03-10 14:58:47 -04:00
51fa86b5c4
refactor: extract run_task_implementation_agent
2025-03-10 14:44:34 -04:00
d8dcc8ca86
refactor: extract run_planning_agent
2025-03-10 14:19:06 -04:00
a437a1e8c3
prompt improvements
2025-03-10 13:46:08 -04:00
4f2c36b958
refactor: extract research agent
2025-03-10 13:33:00 -04:00
2899b5f848
feat(agent_utils.py): add AnthropicCallbackHandler to track token usage and costs for Anthropic models ( #118 )
...
style(agent_utils.py): format imports and code for better readability
refactor(agent_utils.py): standardize model name and cost calculation logic for clarity and maintainability
chore(anthropic_callback_handler.py): create a new file for the AnthropicCallbackHandler implementation and related functions
2025-03-10 07:08:12 -04:00
ddd0e2ae2d
feat(agent_utils.py): add AnthropicCallbackHandler to track token usage and costs for Anthropic models
...
style(agent_utils.py): format imports and code for better readability
refactor(agent_utils.py): standardize model name and cost calculation logic for clarity and maintainability
chore(anthropic_callback_handler.py): create a new file for the AnthropicCallbackHandler implementation and related functions
2025-03-10 01:18:44 -07:00
d194868cff
fix binary detection
2025-03-08 22:36:06 -05:00
053908448f
fix(env_inv.py): handle None values in sorting of installations to prevent errors ( #117 )
2025-03-08 21:03:06 -05:00
d0bf78c5f0
support reasoning assistance in research phase.
2025-03-08 19:32:05 -05:00
0a143e8fff
improve prompts
2025-03-08 17:08:10 -05:00
eaf79fbee5
fix tests; improve prompts
2025-03-08 16:29:17 -05:00
fd772ee369
disable reasoning assistance by default
2025-03-08 15:36:21 -05:00
35f91aa128
add flags to force enable/disable reasoning assistance
2025-03-08 15:36:00 -05:00
ff653c7208
reasoning assistance for implementation tasks
2025-03-08 15:12:13 -05:00
cfb0ec148f
extract thinking processing
2025-03-08 14:48:46 -05:00
8f1e072e20
improve reasoning assistance
2025-03-08 14:23:35 -05:00
2130334837
improve reasoning assistance prompt
2025-03-08 13:50:10 -05:00
d7fb269f4d
improve reasoning assistance prompt
2025-03-08 13:37:38 -05:00
e64c4066a6
implement reasoning-guided workflows
2025-03-08 13:35:06 -05:00
c323098aec
include environment information in prompts
2025-03-08 12:01:40 -05:00
Jose Leon
AI Christianson
Andrew I. Christianson
Ariel Frischer