ReleaseMicrosoftMicrosoftpublished Aug 11, 2025seen 1w

microsoft/debug-gym 1.1.0

microsoft/debug-gym

Open original ↗

Captured source

source ↗
published Aug 11, 2025seen 1wcaptured 1whttp 200method plain

1.1.0

Repository: microsoft/debug-gym

Tag: 1.1.0

Published: 2025-08-11T18:37:21Z

Prerelease: no

Release notes:

What's Changed

  • arxiv url by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/106
  • Readme install from pypi by @matheper in https://github.com/microsoft/debug-gym/pull/108
  • Change credential order in AzureOpenAILLM for proper token retrieval by @matheper in https://github.com/microsoft/debug-gym/pull/110
  • AzureOpenAI Chained Credential by @matheper in https://github.com/microsoft/debug-gym/pull/111
  • Function Calling Syntax by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/109
  • Update human class to use tool calls by @matheper in https://github.com/microsoft/debug-gym/pull/114
  • make sure the working dir is always in sys path by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/115
  • Validate human input by @matheper in https://github.com/microsoft/debug-gym/pull/116
  • Bugfix human test by @matheper in https://github.com/microsoft/debug-gym/pull/118
  • Disable strict mode by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/117
  • Fix env.rewrite_counter by @matheper in https://github.com/microsoft/debug-gym/pull/119
  • Fixed EvalTool being called with extra kwargs when reacting to events by @matheper in https://github.com/microsoft/debug-gym/pull/120
  • Add logs viewer for Froggy by @MarcCote in https://github.com/microsoft/debug-gym/pull/95
  • Fix: Use json.dumps for tool arguments in OpenAILLM by @ShiZhengyan in https://github.com/microsoft/debug-gym/pull/123
  • Tools observations by @matheper in https://github.com/microsoft/debug-gym/pull/121
  • Fix 'utf-8' codec error with surrogate pairs in Unicode strings by @Copilot in https://github.com/microsoft/debug-gym/pull/129
  • Add max_retries parameter to Human class to limit terminal read attempts by @Copilot in https://github.com/microsoft/debug-gym/pull/126
  • Improve test coverage report and default pytest configs by @matheper in https://github.com/microsoft/debug-gym/pull/98
  • Clean up pytest.ini, partially reverting #98 by @matheper in https://github.com/microsoft/debug-gym/pull/132
  • Remove current file by @matheper in https://github.com/microsoft/debug-gym/pull/127
  • Add start and end args to ViewTool by @matheper in https://github.com/microsoft/debug-gym/pull/133
  • Minor Fixes by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/124
  • Fix kwargs by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/135
  • Fix kwargs for pdb tool by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/136
  • Fix type annotation: tool_call_list should be list not dict by @ShiZhengyan in https://github.com/microsoft/debug-gym/pull/134
  • Enhance Agent logging to include step number and reason for termination by @matheper in https://github.com/microsoft/debug-gym/pull/137
  • Improve visualization by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/138
  • Pdb current frame file by @matheper in https://github.com/microsoft/debug-gym/pull/139
  • Pdb breakpoint handling by @matheper in https://github.com/microsoft/debug-gym/pull/140
  • Refactor llm_api into debug_gym.llms subpackage by @MarcCote in https://github.com/microsoft/debug-gym/pull/142
  • Resolve absolute path from RepoEnv by @matheper in https://github.com/microsoft/debug-gym/pull/144
  • Use better command completion for Human Mode by @MarcCote in https://github.com/microsoft/debug-gym/pull/143
  • Fix: resolve_path and is_editable to account for ignored and read-only files by @matheper in https://github.com/microsoft/debug-gym/pull/145
  • Ignore files from .gitignore by @matheper in https://github.com/microsoft/debug-gym/pull/146
  • Fix Aider ignore patterns and add tests for path resolution and ignored/read-only files by @matheper in https://github.com/microsoft/debug-gym/pull/147
  • Fix issue resolving env.working_dir by @matheper in https://github.com/microsoft/debug-gym/pull/148
  • Set default RepoEnv.dir_tree_depth to 1 by @matheper in https://github.com/microsoft/debug-gym/pull/150
  • replace unescape by filtering non-utf8 chars in system prompts by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/151
  • Adding SWE-Smith support by @MarcCote in https://github.com/microsoft/debug-gym/pull/122
  • Only load image for instance_id we want to tests by @MarcCote in https://github.com/microsoft/debug-gym/pull/154
  • No eval shortcut by @matheper in https://github.com/microsoft/debug-gym/pull/152
  • Parallel execution by @matheper in https://github.com/microsoft/debug-gym/pull/153
  • Trajectory Filtering by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/141
  • For SWE-Smith, add a new test split distinct from train-789 by @MarcCote in https://github.com/microsoft/debug-gym/pull/156
  • Fix ViewTool handling empty files by @matheper in https://github.com/microsoft/debug-gym/pull/157
  • Improve Retry by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/158
  • Support pickling PDBTool instances by @threewisemonkeys-as in https://github.com/microsoft/debug-gym/pull/166
  • Resolve path mismatch issue raised on macOS by @dkokkotas in https://github.com/microsoft/debug-gym/pull/159
  • Context change for SFT by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/162
  • Fix PDB indentation mismatch in list output context by @Copilot in https://github.com/microsoft/debug-gym/pull/161
  • Fix init obs by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/171
  • Integrating thinking by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/172
  • Show pytest traceback for test failures. by @MarcCote in https://github.com/microsoft/debug-gym/pull/173
  • Rich logger by @matheper in https://github.com/microsoft/debug-gym/pull/170
  • Agents rich progress by @matheper in https://github.com/microsoft/debug-gym/pull/174
  • A set of fixes by @xingdi-eric-yuan in https://github.com/microsoft/debug-gym/pull/175
  • Add get_problem_ids (formerly get_dataset_split) to all benchmark env by @MarcCote in https://github.com/microsoft/debug-gym/pull/176
  • Pin swe-smith version by @matheper in https://github.com/microsoft/debug-gym/pull/181
  • Change to when tool call is on auto parsing by @icwhite in https://github.com/microsoft/debug-gym/pull/182
  • Add memory limit to Docker containers by @matheper in https://github.com/microsoft/debug-gym/pull/183
  • Disable rich live in human mode by @matheper in...

Excerpt shown — open the source for the full document.

Notability

notability 3.0/10

Routine version update of a niche debugging benchmark.