kshitijk4poor 60262f71bd refactor(mcp-oauth): split the refresh fence into acquire/release functions, drop the lock sidecar on remove ...

The SDK drives a refresh as a generator (request yielded from
_refresh_token, response consumed in _handle_refresh_response), so the
fence was a hand-driven @asynccontextmanager: __aenter__ in one method,
__aexit__ in another, generator object stashed on the provider. A plain
`acquire_refresh_fence(path, timeout) -> fd` / `release_refresh_fence(fd)`
pair says what actually happens and leaves nothing half-entered to leak.
The descriptor is opened with os.open at 0600 and closed on every
acquisition failure.

The `_refresh_token` release-on-exception stays: `_refresh_token` is also
reachable outside `async_auth_flow` (tests call it directly), and the
wrapper's finally only covers the generator-driven path.

HermesTokenStorage.remove() now unlinks the `.refresh.lock` sibling so
logout leaves no stray file. It is deliberately NOT added to
_state_paths(): snapshot()/restore(only_if_absent=True) treat any
existing state path as "newer state exists", and a lingering lock file
would silently veto a rollback.

tests/tools/test_mcp_oauth.py: drop the unused `Path` import (`time` is
still used by the socket poll helper).

2026-09-15 10:50:13 +05:30

..

__init__.py

…

conftest.py

fix(computer_use): approval goes through the shared gate; no callback now fails closed

2026-09-13 05:21:02 -07:00

file_ops_fakes.py

…

test_a2a_reason_roundtrip.py

…

test_accretion_caps.py

test(process_registry): trim handle-release tests to invariants, fix prune fixture

2026-09-13 21:04:45 -07:00

test_allowlist_legacy_config.py

fix(approval): recover legacy list values without character grants

2026-09-07 08:13:38 -07:00

test_allowlist_quoted_metachars.py

…

test_annotate_preview_tool.py

…

test_ansi_strip.py

…

test_apply_layout_tool.py

…

test_approval_config_readonly.py

…

test_approval_deny_rules.py

fix(approvals): honor GNU env split escapes and argv0 operands

2026-09-07 14:30:41 -07:00

test_approval_grep_in_substitution.py

…

test_approval_hook_session_id.py

…

test_approval_interrupt.py

…

test_approval_mode_parity.py

…

test_approval_outcome_parity.py

…

test_approval_plugin_hooks.py

…

test_approval_timeout_overflow.py

…

test_approval_windows.py

…

test_approval.py

fix(tools): reload permanent allowlist by replacement

2026-09-12 22:03:19 -07:00

test_approved_command_clean_slate.py

…

test_async_batch_task_failure_notice.py

…

test_async_delegation_fd_leak.py

test(tools): extend the delegation fd-leak injection to the schema replay

2026-09-11 06:24:54 -07:00

test_async_delegation.py

security(state): harden async_delegation's direct state.db writer

2026-09-12 20:43:42 -07:00

test_audio_container.py

…

test_base_environment.py

test(terminal): trim NOPASSWD probe tests to behaviour contracts

2026-09-11 11:19:53 +05:30

test_binary_document_write_guard.py

…

test_blocked_command_guidance.py

…

test_blueprints.py

…

test_bot_dm_payload_cache.py

…

test_bot_failure_reasons.py

…

test_bot_live_owner_delivery.py

fix(bot-mode): carry the relay sender into a live Bot Chat turn

2026-09-10 10:27:07 -07:00

test_bot_mode_canonical_chat_resolution.py

test: trim the Aug-2026 conformance suites to behaviour invariants on main

2026-09-13 21:13:23 -07:00

test_bot_mode_dm.py

fix(tools): resolve hermes CLI beside the interpreter in bot_mode_dm deliveries

2026-09-14 17:04:56 -07:00

test_bot_mode_probe.py

…

test_bot_mode_refusal_contract.py

fix(bot-mode): preserve refusal reasons across local delivery

2026-09-07 08:15:32 -07:00

test_bot_relay_windows_paths.py

…

test_bot_relay.py

fix(bot-mode): strip canonical session env names, keep HERMES_SESSION_* knobs

2026-09-14 21:18:47 +05:30

test_bot_retry_policy.py

…

test_bot_turn_lock.py

…

test_browser_camofox_auth.py

…

test_browser_camofox_ensure_tab.py

…

test_browser_camofox_persistence.py

…

test_browser_camofox_private_page_guard.py

…

test_browser_camofox_state.py

…

test_browser_camofox_timeout.py

…

test_browser_camofox.py

…

test_browser_cdp_override.py

…

test_browser_cdp_tool.py

…

test_browser_chromium_autoinstall.py

…

test_browser_chromium_check.py

…

test_browser_cleanup.py

chore(tests): encoding=utf-8 on write_text in test_browser_cleanup (windows-footgun ratchet)

2026-09-12 08:43:28 -07:00

test_browser_cloud_fallback.py

…

test_browser_cloud_provider_cache.py

…

test_browser_command_timeout_race.py

…

test_browser_computer_use_profile_cache_keys.py

fix(tools): browser_exec and computer_use caches are namespaced by the served profile

2026-09-13 14:41:26 -07:00

test_browser_console_ssrf.py

…

test_browser_console.py

…

test_browser_content_none_guard.py

…

test_browser_eval_ssrf.py

…

test_browser_eval_supervisor_path.py

…

test_browser_extension_router_wiring.py

…

test_browser_extension_router.py

…

test_browser_get_images_ssrf.py

…

test_browser_hardening.py

…

test_browser_headed_mode.py

…

test_browser_homebrew_paths.py

…

test_browser_hybrid_routing.py

…

test_browser_lightpanda_serve.py

…

test_browser_lightpanda.py

…

test_browser_npx_warmup.py

…

test_browser_open_timeout.py

…

test_browser_orphan_reaper.py

…

test_browser_private_page_action_guard.py

…

test_browser_real_profile_pin.py

test(browser): migrate pinned-profile fixtures to real SQLite

2026-09-08 05:10:57 -07:00

test_browser_real_profile.py

fix(browser): bound auth backups without replacing live SQLite files

2026-09-08 05:10:57 -07:00

test_browser_secret_exfil.py

fix(browser): stop refusing credential-named query params on cloud browser/extract backends

2026-09-10 10:35:07 -07:00

test_browser_snapshot_ssrf.py

…

test_browser_snapshot_threshold.py

…

test_browser_ssrf_local.py

…

test_browser_supervisor_healthcheck.py

…

test_browser_supervisor.py

…

test_browser_suspect_recycle.py

…

test_browser_type_redaction.py

…

test_browser_use_cli.py

fix(browser): resolve the Nous gateway from the picker selection, not only use_gateway

2026-09-13 14:41:26 -07:00

test_browser_use_session_expiry.py

…

test_browser_vault.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_budget_config.py

…

test_build_subprocess_env.py

…

test_checkpoint_manager.py

…

test_checkpoint_path_roundtrip.py

fix: round-trip checkpoint path bytes without text translation

2026-09-07 06:00:46 -07:00

test_clarify_gateway.py

…

test_clarify_tool.py

…

test_clawhub_owner_http.py

test: opt the ClawHub owner-HTTP fixture into private URLs for the SSRF-guarded stream

2026-09-12 21:50:30 -07:00

test_clawhub_zip_stream.py

fix(skills): bound streamed ClawHub ZIP downloads (#57571)

2026-09-12 21:50:30 -07:00

test_cli_approval_exec_ask_leak.py

…

test_clipboard.py

…

test_close_preview_tool.py

…

test_code_execution_hermes_home.py

test(tools): trim the salvaged #110632 / #110307 tests to two invariants each

2026-09-14 16:13:51 -07:00

test_code_execution_modes.py

…

test_code_execution_windows_env.py

…

test_code_execution.py

refactor(config): collapse thin _load_config copies onto the canonical readers

2026-09-13 05:09:06 -07:00

test_code_kernel_remote.py

…

test_code_kernel.py

fix: adapt execute_code cell authority to the widened prompt-callback table

2026-09-10 10:35:07 -07:00

test_command_guards.py

…

test_compaction_tool_refresh.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_completed_process_results.py

test(tools): count each tool receipt once and pin the quiet-notify follow-up

2026-09-14 21:18:47 +05:30

test_computer_use_approval_isolation.py

fix(computer_use): approval goes through the shared gate; no callback now fails closed

2026-09-13 05:21:02 -07:00

test_computer_use_capture_routing.py

…

test_computer_use_cua_0_9.py

fix(computer_use): approval goes through the shared gate; no callback now fails closed

2026-09-13 05:21:02 -07:00

test_computer_use_cua_0_10_permissions.py

fix(computer_use): approval goes through the shared gate; no callback now fails closed

2026-09-13 05:21:02 -07:00

test_computer_use_cua_backend_linux.py

…

test_computer_use_cua_macos_identity.py

…

test_computer_use_delivery_ladder.py

fix(computer_use): approval goes through the shared gate; no callback now fails closed

2026-09-13 05:21:02 -07:00

test_computer_use_display_count_guard.py

…

test_computer_use_empty_discovery_diagnosis.py

…

test_computer_use_fullscreen_capture.py

…

test_computer_use_input_target_guard.py

…

test_computer_use_null_pid_windows.py

…

test_computer_use_placeholder_ids.py

…

test_computer_use_screenshot_dedup.py

fix(computer-use): screenshot dedup forgets its last frame at a compaction boundary

2026-09-14 07:21:18 -07:00

test_computer_use_vision_routing.py

…

test_computer_use_zero_bounds.py

…

test_computer_use.py

test(computer-use): element click carries the token from the live schema alone

2026-09-14 17:34:31 -07:00

test_config_null_guard.py

…

test_connector_bridge_wiring.py

Connector code lives in one package, tools/connectors/ (move only; NS-868 prep) (#110368)

2026-09-15 00:41:13 +05:30

test_connector_dispatch_policy.py

Connector code lives in one package, tools/connectors/ (move only; NS-868 prep) (#110368)

2026-09-15 00:41:13 +05:30

test_connector_local_batches.py

Desktop connect runs on the connection operation: one card, no link to the model, no renderer polling (NS-868) (#110574)

2026-09-15 00:41:14 +05:30

test_connector_names.py

Connector code lives in one package, tools/connectors/ (move only; NS-868 prep) (#110368)

2026-09-15 00:41:13 +05:30

test_connector_session_scope.py

Connector code lives in one package, tools/connectors/ (move only; NS-868 prep) (#110368)

2026-09-15 00:41:13 +05:30

test_connectors_contract.py

Desktop connect runs on the connection operation: one card, no link to the model, no renderer polling (NS-868) (#110574)

2026-09-15 00:41:14 +05:30

test_connectors_gateway_client.py

Connector code lives in one package, tools/connectors/ (move only; NS-868 prep) (#110368)

2026-09-15 00:41:13 +05:30

test_connectors_gateway_merge.py

fix(connectors): manage_connections is absent for accounts the portal has not enabled (#111238)

2026-09-14 22:01:32 +00:00

test_connectors_gateway_wire.py

Desktop connect runs on the connection operation: one card, no link to the model, no renderer polling (NS-868) (#110574)

2026-09-15 00:41:14 +05:30

test_connectors_live.py

Desktop connect runs on the connection operation: one card, no link to the model, no renderer polling (NS-868) (#110574)

2026-09-15 00:41:14 +05:30

test_connectors_managed.py

Desktop connect runs on the connection operation: one card, no link to the model, no renderer polling (NS-868) (#110574)

2026-09-15 00:41:14 +05:30

test_connectors_mcp.py

Desktop connect runs on the connection operation: one card, no link to the model, no renderer polling (NS-868) (#110574)

2026-09-15 00:41:14 +05:30

test_connectors_operation.py

Desktop connect runs on the connection operation: one card, no link to the model, no renderer polling (NS-868) (#110574)

2026-09-15 00:41:14 +05:30

test_connectors_tool.py

fix(connectors): manage_connections is absent for accounts the portal has not enabled (#111238)

2026-09-14 22:01:32 +00:00

test_container_cwd_sanitize.py

…

test_credential_files.py

fix(multiplex): a served profile's turn sees its own cwd, approvals, redaction and tool policy

2026-09-11 19:39:12 -07:00

test_credential_pool_env_fallback.py

…

test_cron_approval_mode.py

…

test_cron_prompt_injection.py

…

test_cronjob_run_background.py

fix(cron): don't stamp the next occurrence on an off-tick manual run

2026-09-09 12:17:13 +05:30

test_cronjob_run_delivery_notice.py

…

test_cronjob_run_immediate.py

fix(cron): don't stamp the next occurrence on an off-tick manual run

2026-09-09 12:17:13 +05:30

test_cronjob_tools.py

feat(cron): add resnap action to adopt the current global inference default

2026-09-12 20:57:21 -07:00

test_cross_profile_guard.py

…

test_daemon_pool.py

…

test_daytona_environment.py

…

test_debug_helpers.py

…

test_deferral_fixes.py

fix(tool-search): a query no tool answers returns nothing, not five tools sharing one word (#106676)

2026-09-10 02:21:15 +05:30

test_delegate_apiserver_background.py

fix: persist API delegation units once without waking the model

2026-09-09 10:55:32 -07:00

test_delegate_batch_tag.py

…

test_delegate_batch_validation.py

…

test_delegate_capability_inheritance.py

…

test_delegate_capacity_interrupt.py

fix(delegation): late-attached children take the parent's soft/hard stop kind; dedupe the fallback replay

2026-09-13 21:31:57 -07:00

test_delegate_child_cache_ttl.py

…

test_delegate_child_compression_cap.py

…

test_delegate_child_transcript_release.py

…

test_delegate_composite_toolsets.py

…

test_delegate_control_actions.py

…

test_delegate_cost_footer.py

…

test_delegate_cron_sync_fallback.py

fix: remove automatic session JSON snapshots

2026-09-07 08:08:41 -07:00

test_delegate_fallback_matrix.py

refactor(delegation): resolve the child fallback chain through the canonical normalizer

2026-09-08 02:26:05 +05:30

test_delegate_group_schema.py

fix: hide inactive grouping options from delegation schema

2026-09-08 13:38:18 -07:00

test_delegate_images.py

fix(delegate): forward inline data-URL images to vision children; trim tests to invariants

2026-09-13 21:05:42 -07:00

test_delegate_kanban_isolation.py

test: align delegated-child env tests with retained board routing

2026-09-07 07:10:28 -07:00

test_delegate_late_child_inherits_stop.py

fix(delegate): grandchildren spawned after their orchestrator was stopped now die with it

2026-09-13 20:13:12 -07:00

test_delegate_output_schema.py

…

test_delegate_request_overrides.py

test: exercise output-cap removal across native and child surfaces

2026-09-07 06:15:43 -07:00

test_delegate_subagent_timeout_diagnostic.py

…

test_delegate_summary_budget.py

…

test_delegate_timeout_cleanup.py

…

test_delegate_toolset_scope.py

…

test_delegate.py

fix(delegation): resolve the child fallback chain through the full pin x config matrix

2026-09-08 02:26:05 +05:30

test_delegation_live_log.py

…

test_denial_circuit_breaker.py

…

test_desktop_tools_diet.py

manage_connections covers local MCP servers; setup_mcp leaves the schema (NS-867, PR1) (#109517)

2026-09-15 00:41:13 +05:30

test_desktop_ui.py

…

test_discord_send_message_caption.py

…

test_discord_tool.py

…

test_display_toggles.py

…

test_docker_cgroup_limits.py

…

test_docker_config_migrate.py

fix(config): one bounded backups/config/ dir replaces four config.yaml.bak schemes

2026-09-09 02:36:00 -07:00

test_docker_daemon_redirect.py

…

test_docker_environment.py

…

test_docker_find.py

…

test_docker_network_config.py

…

test_docker_orphan_reaper_integration.py

…

test_docker_rebootstrap_nous_session.py

fix(docker): rebootstrap re-seed temp is randomly named so a stale temp never blocks recovery

2026-09-13 05:07:11 -07:00

test_docker_session_isolation.py

…

test_docker_teardown_drain.py

…

test_dockerfile_immutable_install.py

…

test_dockerfile_node_modules_perms.py

…

test_dockerfile_pid1_reaping.py

…

test_drive_preview_tool.py

…

test_ensure_task_env.py

…

test_env_passthrough.py

fix(gateway,tools): per-profile Yuanbao home, env_passthrough allowlist and Slack ignored-channel guard under multiplex

2026-09-11 15:29:15 -07:00

test_env_probe.py

…

test_execute_code_approval_cluster.py

test: align delegated-child env tests with retained board routing

2026-09-07 07:10:28 -07:00

test_execute_helper_contract.py

fix(execute-code): teach the working helper import contract

2026-09-07 06:12:24 -07:00

test_execution_flag_detection.py

…

test_fal_common.py

test(fal): trim the billing-409 salvage to two invariant tests

2026-09-09 11:46:36 -07:00

test_feishu_tools.py

…

test_file_operations_edge_cases.py

fix(tools): count final unterminated line in read_file total_lines

2026-09-12 20:52:49 -07:00

test_file_operations.py

fix(tools): stop read_file rendering a phantom empty line for newline-terminated files

2026-09-12 20:52:49 -07:00

test_file_ops_cwd_tracking.py

…

test_file_ops_env_not_ready.py

…

test_file_ops_single_roundtrip.py

fix(tools): read_file line accounting — count unterminated final lines, drop the phantom trailing line

2026-09-12 20:52:49 -07:00

test_file_read_guards.py

fix(tools): profile-scoped checkpoint/snapshot paths, tool caches, TZ and schema paths under multiplex

2026-09-11 15:44:00 -07:00

test_file_staleness.py

…

test_file_state_registry.py

…

test_file_sync_back.py

…

test_file_sync_perf.py

…

test_file_sync_sigint.py

…

test_file_sync_upload_snapshot.py

fix(file-sync): hash the uploaded snapshot instead of mutable host files

2026-09-07 06:02:41 -07:00

test_file_sync.py

test: use explicit UTF-8 in file sync fixtures

2026-09-07 06:02:41 -07:00

test_file_tools_container_config.py

…

test_file_tools_cwd_resolution.py

…

test_file_tools_live.py

…

test_file_tools_plugin_container_cwd.py

…

test_file_tools_tilde_profile.py

…

test_file_tools.py

…

test_file_write_safety.py

test(tools): trim the salvaged #110632 / #110307 tests to two invariants each

2026-09-14 16:13:51 -07:00

test_file_write_surrogate_roundtrip.py

…

test_find_shell.py

test(tools): pin the Git Bash probe's detached stdin in the existing probe test

2026-09-08 02:26:39 +05:30

test_finite_delegation_outcomes.py

fix: join delegated work before finite chat exits

2026-09-10 05:02:41 -07:00

test_focus_pane_tool.py

…

test_force_dangerous_override.py

…

test_fuzzy_match.py

fix(tools): make the empty/whitespace old_string rejection actionable

2026-09-12 21:55:07 -07:00

test_gateway_cwd_contract.py

…

test_generation_source_confinement.py

…

test_gnu_long_option_abbreviation_bypass.py

…

test_hardline_blocklist.py

…

test_hardline_escaped_quote_state.py

fix(approval): judge shell quoting on the raw command, not the escape-stripped one

2026-09-14 09:19:25 -07:00

test_heartbeat_stale_thresholds.py

…

test_hermes_subprocess_env.py

test: align delegated-child env tests with retained board routing

2026-09-07 07:10:28 -07:00

test_hidden_dir_filter.py

…

test_homeassistant_tool.py

…

test_hook_output_spill.py

…

test_hub_lock_non_utf8.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_image_generate_schema.py

…

test_image_generation_artifacts.py

…

test_image_generation_env.py

…

test_image_generation_image_to_image.py

feat(image-gen): Kling Image v3 in the FAL catalog (t2i + singular-key i2i)

2026-09-13 21:03:11 -07:00

test_image_generation_interrupt.py

…

test_image_generation_plugin_dispatch.py

…

test_image_generation.py

feat(image-gen): Kling Image v3 in the FAL catalog (t2i + singular-key i2i)

2026-09-13 21:03:11 -07:00

test_image_source.py

test(vision): import HEIC helpers from the defining sibling module

2026-09-13 09:18:54 -07:00

test_init_session_cwd_respect.py

…

test_interrupt.py

…

test_interrupted_command_cwd.py

…

test_kanban_comment_injection.py

…

test_kanban_descendant_scope.py

fix: keep Kanban worker scope out of descendant processes

2026-09-07 07:10:28 -07:00

test_kanban_provenance.py

fix(kanban): preserve durable origins for worker-created tasks

2026-09-07 14:16:57 -07:00

test_kanban_redaction.py

…

test_kanban_tools.py

fix(kanban): an explicit scratch workspace never inherits the board's project

2026-09-09 12:18:58 -07:00

test_kanban_toolset_opt_in.py

fix(kanban): drop the TUI empty-selection change and refit tests to the opt-in contract

2026-09-12 12:32:55 -07:00

test_launchctl_guard_diagnostic.py

fix: explain the launchctl registration restriction without inventing KeepAlive

2026-09-07 06:05:51 -07:00

test_lazy_deps_durable_target.py

…

test_lazy_deps_managed.py

…

test_lazy_deps.py

…

test_line_ending_preservation.py

…

test_llm_content_none_guard.py

…

test_local_background_child_hang.py

…

test_local_cwd_permission_fallback.py

…

test_local_env_blocklist.py

…

test_local_env_cwd_recovery.py

…

test_local_env_relative_cwd.py

…

test_local_env_session_leak.py

…

test_local_env_windows_msys.py

…

test_local_interrupt_cleanup.py

…

test_local_setsid_descendant_sweep.py

…

test_local_shell_init.py

…

test_local_temp_dir.py

…

test_local_tempdir.py

…

test_macos_protected_search.py

…

test_managed_browserbase_and_modal.py

test(tools): keep unmocked package dependencies importable

2026-09-07 14:10:46 -07:00

test_managed_media_gateways.py

test(fal): trim the billing-409 salvage to two invariant tests

2026-09-09 11:46:36 -07:00

test_managed_modal_environment.py

…

test_managed_tool_gateway.py

feat(desktop): connector cards that wait for the sign-in, and a guided first launch that holds together (#108292)

2026-09-12 10:00:06 +05:30

test_mcp_bridge_single_failure.py

…

test_mcp_cancelled_error_propagation.py

…

test_mcp_capability_gating.py

…

test_mcp_cimd.py

…

test_mcp_circuit_breaker.py

fix(mcp): breaker opened by tool errors says "rejected", not "unreachable"

2026-09-12 09:05:56 -07:00

test_mcp_client_cert.py

fix(mcp): cap HTTP/SSE response bodies before SDK parse

2026-09-13 20:45:25 -07:00

test_mcp_config_whitespace_warning.py

…

test_mcp_dashboard_oauth.py

…

test_mcp_death_supervisor.py

…

test_mcp_device_flow.py

fix: restore prior device OAuth state if persistence fails

2026-09-07 08:22:35 -07:00

test_mcp_discovery_cross_process.py

…

test_mcp_dynamic_discovery.py

…

test_mcp_elicitation.py

test: e2e conftest imports GatewayRunner at module level; elicitation test docstring matches the default thunk

2026-09-07 22:47:33 +05:30

test_mcp_empty_error_message.py

…

test_mcp_failure_classification.py

…

test_mcp_http_body_cap.py

test: drop upstream product reference from body-cap test docstring

2026-09-13 20:45:25 -07:00

test_mcp_identity_header.py

…

test_mcp_image_content.py

…

test_mcp_initial_connect_shutdown.py

…

test_mcp_invalid_url.py

…

test_mcp_lazy_start.py

…

test_mcp_list_pagination.py

fix: keep pagination signature inspection local and preserve exact decoder error

2026-09-07 07:04:52 -07:00

test_mcp_loop_profile_override.py

…

test_mcp_multiplex_connection_keys.py

fix(mcp): supports_parallel_tool_calls is per profile, not per server name

2026-09-13 15:41:01 -07:00

test_mcp_npx_cached_bin.py

…

test_mcp_oauth_bidirectional.py

…

test_mcp_oauth_cold_load_expiry.py

…

test_mcp_oauth_integration.py

…

test_mcp_oauth_issuer_binding.py

feat(mcp): bind stored OAuth refresh tokens to their issuer

2026-09-13 20:43:33 -07:00

test_mcp_oauth_manager.py

refactor(mcp-oauth): split the refresh fence into acquire/release functions, drop the lock sidecar on remove

2026-09-15 10:50:13 +05:30

test_mcp_oauth_metadata.py

…

test_mcp_oauth_stdin_console.py

fix(mcp): unattended paths never open browser OAuth; gateway follows mcp_servers edits

2026-09-13 06:29:12 -07:00

test_mcp_oauth_user_agent.py

…

test_mcp_oauth.py

refactor(mcp-oauth): split the refresh fence into acquire/release functions, drop the lock sidecar on remove

2026-09-15 10:50:13 +05:30

test_mcp_parked_probe_no_browser.py

fix(mcp): unattended paths never open browser OAuth; gateway follows mcp_servers edits

2026-09-13 06:29:12 -07:00

test_mcp_parked_self_probe.py

…

test_mcp_poll_loop_oom_integration.py

…

test_mcp_preflight_content_type.py

…

test_mcp_probe.py

…

test_mcp_protocol_negotiation.py

…

test_mcp_rapid_drop_budget.py

…

test_mcp_reconcile_with_config.py

fix(mcp): reconcile chore lives in run_profile_reconcile; prune lazy + mid-connect servers

2026-09-13 06:29:12 -07:00

test_mcp_reconnect_log_hygiene.py

…

test_mcp_reconnect_retry_reset.py

…

test_mcp_reconnect_signal.py

…

test_mcp_register_wakes_stale.py

…

test_mcp_resource_content.py

…

test_mcp_result_size_limit.py

…

test_mcp_schema_cache_ttl.py

…

test_mcp_schema_cache.py

…

test_mcp_server_log_notifications.py

…

test_mcp_sse_fallback.py

fix(mcp): widen the SSE fallback trigger and harden its guards (salvage #53764)

2026-09-12 21:05:29 -07:00

test_mcp_sse_transport.py

…

test_mcp_stability.py

…

test_mcp_stdio_children_dead.py

…

test_mcp_stdio_encoding_handler.py

…

test_mcp_stdio_fastfail_reconnect.py

fix(mcp): classify an SDK-first transport close after dispatch as an ambiguous stdio death

2026-09-09 11:27:29 -07:00

test_mcp_stdio_init_timeout.py

…

test_mcp_streamable_http_arity.py

…

test_mcp_structured_content.py

…

test_mcp_timeout_resolution.py

…

test_mcp_tool_401_handling.py

…

test_mcp_tool_issue_948.py

…

test_mcp_tool_session_expired.py

fix(mcp): recovered application errors keep the breaker strike

2026-09-12 08:23:28 -07:00

test_mcp_tool.py

fix(mcp): clamp generated MCP tool names to 64 chars

2026-09-13 20:44:29 -07:00

test_mcp_transport_group_reconnect.py

…

test_mcp_trust_gating.py

…

test_mcp_utility_capability_gating.py

…

test_media_caption_split.py

…

test_memory_tool_import_fallback.py

…

test_memory_tool_schema.py

…

test_memory_tool.py

fix: keep the ssh_access id, word-bound the verb gate, collapse tests to two invariants

2026-09-14 16:16:08 -07:00

test_microsoft_graph_auth.py

…

test_microsoft_graph_client.py

…

test_modal_bulk_upload.py

…

test_modal_sandbox_fixes.py

…

test_modal_snapshot_isolation.py

test(tools): keep unmocked package dependencies importable

2026-09-07 14:10:46 -07:00

test_model_tools.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_multiplex_tool_cache_scope.py

fix(multiplex): key tool-side and agent-side memos by profile home

2026-09-12 01:35:05 -07:00

test_multiplex_tool_credential_scope.py

fix(multiplex): tool and memory-provider env reads stay inside the routed profile

2026-09-11 15:26:46 -07:00

test_multiplex_tool_memo_scope.py

fix(tools): profile-scoped checkpoint/snapshot paths, tool caches, TZ and schema paths under multiplex

2026-09-11 15:44:00 -07:00

test_multiplex_turn_parity.py

fix(multiplex): a served profile's turn sees its own cwd, approvals, redaction and tool policy

2026-09-11 19:39:12 -07:00

test_notify_on_complete.py

…

test_oneshot_completion_linger.py

fix(tools): one-shot linger waits for a completion that is mid-publish

2026-09-14 20:44:56 -07:00

test_open_preview_tool.py

…

test_osv_check.py

Port from aaif-goose/goose#11466: recognize Windows package-runner shims in OSV malware preflight

2026-09-13 20:46:22 -07:00

test_parse_env_var.py

…

test_patch_already_applied.py

…

test_patch_failure_tracking.py

…

test_patch_multimatch_locations.py

…

test_patch_parser.py

fix(tools): reject V4A Add File onto an existing path

2026-09-12 21:13:40 -07:00

test_patch_v4a_gate.py

…

test_patch_ws_diagnosis.py

…

test_permanent_allowlist_reconcile.py

test(approval): reconcile fixture uses the per-home baseline map

2026-09-12 22:03:19 -07:00

test_plugin_compat_warning.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_plugin_guard_block_reason.py

fix(plugin): identify critical findings in install blocks

2026-09-13 14:43:04 -07:00

test_plugin_guard.py

test: trim prose/comment guard coverage to invariants; drop the unused CSS comment prefix

2026-09-14 16:12:07 -07:00

test_plugin_skills.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_pr_6656_regressions.py

…

test_pre_transcription_hook.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_process_registry_list_exit.py

fix(process): list refreshes no longer leave exited children running

2026-09-12 22:06:03 -07:00

test_process_registry_windows_live.py

…

test_process_registry_write_stdin_surrogates.py

…

test_process_registry.py

test(process_registry): trim handle-release tests to invariants, fix prune fixture

2026-09-13 21:04:45 -07:00

test_process_schema_diet.py

test: derive process_manage verbs from the handler table; widen aux-cancellation start-up bound

2026-09-07 12:50:29 -07:00

test_process_wait_clarity.py

fix(tools): steer/redirect releases a blocking process_manage wait (port of MoonshotAI/kimi-code#3697)

2026-09-12 21:25:00 -07:00

test_pty_query_responder.py

Port from openai/codex#41436: answer blocking terminal queries in background PTY sessions

2026-09-12 22:09:19 -07:00

test_react_to_message_tool.py

…

test_read_binary_type_disclosure.py

…

test_read_extract.py

…

test_read_file_schema_gating.py

…

test_read_file_utf8_binary_regression.py

…

test_read_loop_detection.py

…

test_read_past_eof_note.py

…

test_read_preview_tool.py

…

test_read_shell_line_clamp.py

…

test_read_special_file_guard.py

…

test_read_unicode_filename_retry.py

…

test_read_window_tool.py

…

test_refresh_agent_mcp_tools.py

test(multiplex): invariants for the residue fixes; retire allowlist tests

2026-09-11 19:50:46 -07:00

test_registry.py

Connector code lives in one package, tools/connectors/ (move only; NS-868 prep) (#110368)

2026-09-15 00:41:13 +05:30

test_removed_backend_migration.py

…

test_request_tool_approval.py

fix(approval): approvals.mode off bypasses the shared action gate (computer_use prompts)

2026-09-13 19:58:33 -07:00

test_resolve_path.py

…

test_restored_delegation_ownership.py

…

test_rollback_all_directories.py

…

test_safe_getcwd_permission_error.py

…

test_sandbox_failure_hints.py

fix(execute-code): teach the working helper import contract

2026-09-07 06:12:24 -07:00

test_schema_boolean_required.py

fix(schemas): preserve required intent without invalid boolean flags

2026-09-07 08:14:51 -07:00

test_schema_sanitizer.py

…

test_search_auto_multiline.py

…

test_search_budget_truncation.py

…

test_search_error_guard.py

…

test_search_files_cpu_windows.py

…

test_search_files_engine_selection.py

…

test_search_giant_line_containment.py

…

test_search_hidden_dirs.py

test(search): use neutral visible-content fixture text

2026-09-07 14:10:46 -07:00

test_search_native_rg.py

…

test_search_zero_match_and_multipath.py

…

test_self_repo_guard.py

…

test_send_message_cross_loop.py

…

test_send_message_missing_platforms.py

…

test_send_message_multiplex_profile_adapter.py

fix(gateway): secondary-profile send_message, notices and /loop wakeups go out via their own bot

2026-09-10 18:17:22 -07:00

test_send_message_plugin_extensibility.py

…

test_send_message_react.py

…

test_send_message_relay_target_authz.py

fix(relay): authorize send_message targets and surface egress declines (P5) (#99220)

2026-09-08 17:51:35 +10:00

test_send_message_slack.py

…

test_send_message_target_parse.py

…

test_send_message_telegram_proxy.py

revert: remove Collective Wisdom V1 (#94266)

2026-09-11 11:54:49 -07:00

test_send_message_tool.py

revert: remove Collective Wisdom V1 (#94266)

2026-09-11 11:54:49 -07:00

test_session_cwd_store.py

…

test_session_search.py

fix(session_search): a bare session id never reads another profile's state.db

2026-09-11 06:24:11 -07:00

test_shared_container_task_id.py

…

test_shell_bypass_denylist.py

…

test_signal_media.py

…

test_single_query_approval_mode.py

…

test_singularity_persistent_overlay.py

…

test_singularity_preflight.py

…

test_skill_bundle_provenance.py

…

test_skill_create_dir.py

…

test_skill_env_passthrough.py

fix(gateway,tools): per-profile Yuanbao home, env_passthrough allowlist and Slack ignored-channel guard under multiplex

2026-09-11 15:29:15 -07:00

test_skill_improvements.py

…

test_skill_ledger.py

…

test_skill_linter.py

…

test_skill_manage_batch.py

…

test_skill_manage_schema_diet.py

revert: remove Collective Wisdom V1 (#94266)

2026-09-11 11:54:49 -07:00

test_skill_manager_tool.py

…

test_skill_provenance.py

fix(review): keep /refine under the background_review origin; attendedness is its own flag

2026-09-09 12:19:13 +05:30

test_skill_size_limits.py

…

test_skill_usage.py

revert: remove Collective Wisdom V1 (#94266)

2026-09-11 11:54:49 -07:00

test_skill_view_dedup.py

revert: remove Collective Wisdom V1 (#94266)

2026-09-11 11:54:49 -07:00

test_skill_view_path_check.py

…

test_skill_view_traversal.py

…

test_skillevaluator_scan.py

…

test_skills_ast_audit.py

…

test_skills_guard_agent_config.py

test: assert skills-guard scanner version is past v1, not pinned to v2

2026-09-14 14:41:50 -07:00

test_skills_guard.py

fix(skills-guard): socat only flags a reverse shell when an address spec follows

2026-09-14 21:55:24 -07:00

test_skills_hub_browse_sh.py

…

test_skills_hub_clawhub.py

fix(skills): bound streamed ClawHub ZIP downloads (#57571)

2026-09-12 21:50:30 -07:00

test_skills_hub.py

feat(skills-hub): tap K-Dense and OpenScience scientific skills under one "science" bucket

2026-09-12 07:54:04 -07:00

test_skills_list_modified_diff.py

…

test_skills_runtime_cache.py

…

test_skills_sync_client.py

revert: remove Collective Wisdom V1 (#94266)

2026-09-11 11:54:49 -07:00

test_skills_sync.py

…

test_skills_tool_discovery_cache.py

…

test_skills_tool_profile_scope.py

…

test_skills_tool.py

revert: remove Collective Wisdom V1 (#94266)

2026-09-11 11:54:49 -07:00

test_slack_send_message_media.py

…

test_slash_confirm.py

…

test_smart_approval_injection.py

…

test_smart_approval_policy.py

…

test_snapshot_multiline_session_env_injection.py

…

test_snapshot_session_id_leak.py

…

test_spill_safety.py

…

test_spotify_client.py

…

test_ssh_bulk_upload.py

…

test_ssh_environment.py

…

test_stage2_hook_api_server_keygen.py

…

test_stage2_hook_seed_one_symlinks.py

…

test_stage2_hook_symlink_chown.py

…

test_startup_latency_regressions.py

test: cover adapter-wrapped probe stubs and repeat-probe stability

2026-09-13 11:22:08 -07:00

test_strict_provider_selection.py

…

test_stt_cloud_trim.py

…

test_stt_default_language.py

…

test_stt_idle_unload.py

…

test_stt_language_resolution.py

…

test_stt_silence_hallucinations.py

…

test_subagent_process_handoff.py

feat(delegation): report a child's exited-but-unread notify processes to the parent

2026-09-07 12:50:29 -07:00

test_subagent_steer.py

test(contracts): tests mirror tui_gateway/; the runtime-artifact spoof test asserts the new 4000

2026-09-14 06:12:19 -07:00

test_subagent_sudo_prompt.py

refactor(terminal): probe NOPASSWD only when a prompt would fire; drop the host-only probe

2026-09-11 11:19:53 +05:30

test_subagent_worktree.py

…

test_subprocess_home_isolation.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_subprocess_stdin_guard.py

…

test_subprocess_utf8_encoding.py

…

test_symlink_prefix_confusion.py

…

test_sync_back_backends.py

…

test_telegram_send_message_caption.py

revert: remove Collective Wisdom V1 (#94266)

2026-09-11 11:54:49 -07:00

test_terminal_bounded_execute.py

…

test_terminal_compound_background.py

…

test_terminal_config_env_sync.py

…

test_terminal_cwd_echo.py

…

test_terminal_degraded_mode.py

…

test_terminal_env_bridge.py

fix(tools): never ambient-bridge TERMINAL_* under a profile home override

2026-09-11 15:29:15 -07:00

test_terminal_environment_recreated_notice.py

Port from lobehub/lobehub#19329: surface environment recreation in terminal tool results

2026-09-12 20:55:04 -07:00

test_terminal_error_redaction.py

…

test_terminal_exit_semantics.py

…

test_terminal_foreground_timeout_cap.py

…

test_terminal_heredoc_background_guard.py

…

test_terminal_hints.py

…

test_terminal_none_command_guard.py

…

test_terminal_output_transform_hook.py

…

test_terminal_requirements.py

…

test_terminal_scope_multiplex.py

fix(tui): bind launch-profile terminal scope once multiplexing is active

2026-09-11 15:29:15 -07:00

test_terminal_self_repo_guard.py

…

test_terminal_signal_exit.py

…

test_terminal_task_cwd.py

…

test_terminal_timeout_output.py

…

test_terminal_tool_exception_redaction.py

…

test_terminal_tool_pty_fallback.py

…

test_terminal_tool_requirements.py

fix(tools): uncached check_fn probes classify UnscopedSecretError from the live scope

2026-09-14 16:13:51 -07:00

test_terminal_tool.py

test(terminal): trim NOPASSWD probe tests to behaviour contracts

2026-09-11 11:19:53 +05:30

test_terminal_truncation_spill.py

fix(tools): profile-scoped checkpoint/snapshot paths, tool caches, TZ and schema paths under multiplex

2026-09-11 15:44:00 -07:00

test_terminal_yield_to_background.py

…

test_termux_api_detection.py

…

test_threaded_process_handle.py

…

test_threat_patterns.py

fix(threat-scanner): gate ssh_access on every mutating verb, not just copy verbs

2026-09-14 16:16:08 -07:00

test_timeout_transport_drain.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_tip_tool.py

…

test_tirith_security.py

…

test_todo_nested.py

…

test_todo_schema_diet.py

…

test_todo_tool_type_coercion.py

…

test_todo_tool.py

…

test_tool_backend_helpers.py

…

test_tool_output_limits.py

…

test_tool_output_truncate.py

refactor(agent): one Retry-After parser and one reset-grammar table feed every retry wait

2026-09-13 05:09:43 -07:00

test_tool_result_storage.py

fix(multiplex): a served profile's turn sees its own cwd, approvals, redaction and tool policy

2026-09-11 19:39:12 -07:00

test_tool_search_context_provider.py

…

test_tool_search_multiquery.py

Connector tools (Gmail, Linear, Notion, ...) are searchable and callable through tool_search for signed-in Nous users (#106842)

2026-09-10 02:21:16 +05:30

test_tool_search.py

feat(tool_search): long hunts for nonexistent tools now return no results instead of incidental matches

2026-09-13 21:04:08 -07:00

test_toolset_distributions.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_toolsets.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_tour_tool.py

…

test_transcription_command_providers.py

…

test_transcription_deepinfra.py

…

test_transcription_dotenv_fallback.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_transcription_plugin_dispatch.py

…

test_transcription_tools.py

…

test_transcription.py

…

test_translate_execute_context.py

fix(threats): keep unrelated role prose in context files

2026-09-07 08:09:20 -07:00

test_truncation_footer_sandbox_paths.py

…

test_tts_command_providers.py

…

test_tts_container_repair.py

…

test_tts_deepinfra.py

…

test_tts_dotenv_fallback.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_tts_gemini.py

…

test_tts_instructions.py

…

test_tts_kittentts.py

…

test_tts_lifecycle_leases.py

…

test_tts_long_form_chunking.py

…

test_tts_macos_output.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_tts_max_text_length.py

…

test_tts_minimax_region.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_tts_mistral.py

…

test_tts_model_cache_lru.py

…

test_tts_openai_config.py

…

test_tts_opus_routing.py

…

test_tts_output_dir_profile_scope.py

…

test_tts_output_timestamp.py

…

test_tts_path_traversal.py

…

test_tts_piper.py

…

test_tts_plugin_dispatch.py

…

test_tts_prepare_spoken.py

…

test_tts_provider_base_urls.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_tts_pythonpath_fallback.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_tts_response_body_cap.py

…

test_tts_speed.py

…

test_tts_streaming_e2e.py

…

test_tts_streaming.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_tts_text_normalize.py

…

test_tts_xai_speech_tags.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_unicode_tag_strip.py

…

test_url_safety.py

…

test_utf16_read.py

…

test_vercel_sandbox_environment.py

…

test_video_analyze.py

…

test_video_generate_schema.py

feat(video): OpenRouter backend covers every model on the live video catalog

2026-09-12 13:44:52 -07:00

test_video_generation_dispatch.py

…

test_video_generation_dynamic_schema.py

…

test_video_generation_tool_surface_matrix.py

test(video-gen): keep the i2v-only guard real once every family is dual-modality

2026-09-13 21:40:26 -07:00

test_vision_native_fast_path.py

…

test_vision_region.py

…

test_vision_scale_disclosure.py

…

test_vision_tools.py

fix(tools): stop check_fns swallowing resolver crashes into "returned False"

2026-09-13 12:35:36 -07:00

test_voice_cli_integration.py

…

test_voice_client_config.py

…

test_voice_credential_pool_resolution.py

…

test_voice_max_recording_seconds.py

test: make tests/ mirror the source tree; drop issue numbers from filenames

2026-09-13 09:18:02 -07:00

test_voice_mode_playback_env_scrub.py

…

test_voice_mode.py

fix(voice): retry a timed-out audio input stream start once

2026-09-13 14:39:29 -07:00

test_voice_stop_phrase.py

…

test_voice_thinking_sound.py

…

test_voice_tts_echo_guard.py

…

test_voice_wsl_pipewire.py

…

test_wake_word.py

…

test_watch_patterns.py

…

test_web_extract_robustness.py

…

test_web_extract_timeout.py

fix(web): cap web_extract provider dispatch with a wall-clock timeout (salvage #57180)

2026-09-12 21:30:38 -07:00

test_web_keyless_fallback.py

…

test_web_keyless_rescue.py

…

test_web_providers_brave_free.py

…

test_web_providers_ddgs.py

…

test_web_providers_searxng.py

…

test_web_providers_xai.py

…

test_web_providers.py

…

test_web_result_cache.py

…

test_web_tools_config.py

…

test_web_tools_dict_urls.py

…

test_web_tools_perplexity.py

feat(web): send Hermes identity headers on Perplexity requests

2026-09-14 13:04:39 -07:00

test_web_tools_tavily.py

…

test_web_tools_truncate.py

…

test_whatsapp_send_message_media.py

…

test_windows_agent_loop_papercuts.py

…

test_windows_compat.py

…

test_windows_native_support.py

…

test_working_diff.py

…

test_write_approval.py

…

test_write_deny.py

…

test_write_file_rewrite_hint.py

…

test_write_file_syntax_gate.py

…

test_write_verification.py

…

test_x_search_tool.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_xai_http_credentials.py

refactor(secrets): drop scope-aware env shims; runtime_provider and the voice/xai tools read the canonical getters

2026-09-13 05:07:50 -07:00

test_xai_http_storage.py

…

test_yolo_mode.py

…

test_zombie_process_cleanup.py

…