Murali Chillakuru 085e005d92 security: remove shell=True from ReAct agent bash tool ...

Parse the command with shlex.split and run with shell=False, and restrict the executable to a python/python3 allow-list. This removes shell metacharacter injection (the benchmark only needs Python to manipulate spreadsheets). Adds tests for the allow-list gate and non-interpretation of shell metacharacters.

2026-07-25 18:41:23 -04:00

..

__init__.py

…

test_alfworld_paths.py

…

test_azure_openai_compat.py

fix(sleep): harden OpenAI compatibility boundaries

2026-07-14 09:17:18 +00:00

test_claude_backend_tempdir.py

fix(claude): use cleanup-tolerant temporary directory

2026-07-14 08:46:54 +00:00

test_codex_optimizer_backend.py

fix(model): isolate Codex token tracking

2026-07-16 10:02:21 +00:00

test_consolidate_split.py

fix(sleep): keep all-test batches out of consolidate train/val

2026-07-24 00:05:34 +08:00

test_cursor_exec_backend.py

Guard Cursor file-edit rollouts with sandboxing

2026-07-20 21:15:04 +01:00

test_devin_plugin.py

fix(devin): make SessionEnd installation non-destructive

2026-07-21 18:02:43 +00:00

test_gate.py

…

test_handoff_backend.py

…

test_json_utils.py

perf(json): keep array scan linear on malformed output

2026-07-14 08:46:54 +00:00

test_materialize_searchqa.py

…

test_mcp_schema.py

Slim down surface for this integration

2026-07-19 22:37:20 +01:00

test_openai_compatible_backend.py

…

test_plugin_sync.py

Disable unverified Cursor Sleep tool replay

2026-07-20 21:14:39 +01:00

test_qwen_backend.py

…

test_react_agent_no_shell.py

security: remove shell=True from ReAct agent bash tool

2026-07-25 18:41:23 -04:00

test_run_sleep_fallback.py

…

test_runners.Tests.ps1

fix(windows): resolve control flow and path quoting in scheduler task execution

2026-07-15 14:54:27 +05:30

test_scheduler_windows.py

fix(windows): resolve control flow and path quoting in scheduler task execution

2026-07-15 14:54:27 +05:30

test_scheduler.py

…

test_scoring.py

…

test_searchqa_rollout_failfast.py

…

test_semantic_density.py

…

test_skill_aware_reflection.py

…

test_sleep_engine.py

fix(sleep): harden evidence log privacy

2026-07-21 17:25:52 +00:00

test_sleep_evidence.py

fix(sleep): harden evidence log privacy

2026-07-21 17:25:52 +00:00

test_types.py

…

test_webui_env_preflight.py

…