chelsealong 316d52faf2 fix(batch_runner): write a discard tombstone so resume skips no-reasoning prompts ...

The no-reasoning discard branch in _process_batch_worker continued
before writing any JSONL row, so run(resume=True) — which filters
solely via _scan_completed_prompts_by_content over batch_*.jsonl —
never saw discarded prompts and re-ran them at full cost on every
resume. Write a tombstone row on discard, exclude tombstones from the
trajectories.jsonl merge, and report discarded_no_reasoning in
final statistics.

Salvaged from #93542.
Fixes #93527

2026-08-24 03:21:07 -07:00

..

acp

fix(approval): stop the CLI and ACP offering a scope the protected gate discards

2026-08-23 17:45:47 -05:00

acp_adapter

fix(security): redact terminal exception results and ACP stderr logs (#77484)

2026-08-08 04:19:49 -07:00

agent

test(model_metadata): lock in max_tokens last-resort fallback + cache self-heal

2026-08-24 03:20:57 -07:00

ci

fix(ci): declare the rust lane on the detect job, and test that wiring

2026-08-20 11:46:30 -05:00

cli

fix(approval): stop the CLI and ACP offering a scope the protected gate discards

2026-08-23 17:45:47 -05:00

computer_use

test(cua): pin PATH-preservation contract, not byte equality

2026-08-23 18:27:15 -07:00

conformance

…

cron

fix(cron): prevent empty payload loop and protect against blank name overwrite

2026-08-24 15:47:42 +05:30

dashboard

…

docker

fix(docker): update runtime tests and docs for the entrypoint dispatcher

2026-08-01 10:52:34 -07:00

e2e

…

fakes

…

fixtures

fix(desktop): preserve turn-elapsed timer across session switches

2026-08-15 13:35:32 -07:00

gateway

fix(gateway): resolve PairingStore's default pairing dir lazily, not at import time

2026-08-24 03:20:46 -07:00

hermes_cli

fix(cli): add --reasoning to both top-level value-flag sets

2026-08-24 03:20:37 -07:00

hermes_state

fix(tui): sweep orphaned tui/desktop/subagent session rows at gateway startup

2026-08-23 18:58:40 -07:00

honcho_plugin

fix(honcho): gate memory-file migration on the declared owner

2026-08-13 23:43:15 +05:30

install

test(install): prove updating from a release reaches this commit

2026-08-04 17:36:26 -04:00

integration

…

manual

…

monitoring

fix(otlp): span exporter now inherits configured resource_attributes

2026-08-08 18:11:02 -07:00

openviking_plugin

fix(openviking): integrate reliability and configuration hardening

2026-08-03 20:35:47 +05:30

plugins

fix(kanban): reuse event stream database connection

2026-08-23 15:27:24 -07:00

providers

fix(nous): stop dropping "thinking off" on Portal models that can honor it

2026-08-19 22:14:56 -05:00

relay

fix(relay): prompt-lifecycle acks are fire-and-forget — awaiting them ON the read loop self-deadlocked the transport

2026-08-20 19:56:01 -07:00

run_agent

fix: align cache-policy pre-gate identity with the capability matcher

2026-08-24 14:58:42 +05:30

scripts

test(ci): close the pytest-wrapper gap for check-windows-footguns.py

2026-08-13 02:38:28 -07:00

secret_sources

fix(plugins): isolate ownership by profile

2026-08-12 19:13:32 -07:00

skills

feat(skill): route hermes-agent's unknown-feature questions to llms.txt

2026-08-19 12:41:45 -05:00

state

fix(state): reap only proven database holders

2026-08-23 20:01:41 -07:00

tools

fix(tools): widen the command-position anchor to the whole hardline class

2026-08-24 03:20:14 -07:00

tui_gateway

fix(desktop): probe half-open gateway socket on wake and reconnect

2026-08-24 03:14:19 -07:00

verify

Integrate verify subsystem with the existing verification stack

2026-08-07 10:11:05 -07:00

website

fix(docs): index every docs page in llms.txt, not a hand-picked 98

2026-08-19 12:41:45 -05:00

__init__.py

…

conftest.py

fix: 7 GitHub-adjacent tests no longer fail on developer machines

2026-08-19 01:19:28 -07:00

run_interrupt_test.py

…

test_89315_replace_ownership_guard.py

fix(gateway): prove --replace ownership from the bound pid record

2026-08-24 15:37:22 +05:30

test_account_usage.py

…

test_atomic_replace_symlinks.py

test(utils): cover contended Windows renames on the Windows CI lane

2026-08-12 17:25:30 -05:00

test_atomic_write_text_metadata.py

fix(utils): tighten create_mode semantics and close the yaml 0600 transit window

2026-08-06 05:00:17 +05:30

test_audio_playback_guard.py

fix: use handle_request for voice.toggle in audio guard test

2026-08-13 22:44:23 +05:30

test_background_review_list_shapes.py

…

test_background_review_session_isolation.py

…

test_base_url_hostname.py

…

test_batch_runner_checkpoint.py

fix(batch_runner): write a discard tombstone so resume skips no-reasoning prompts

2026-08-24 03:21:07 -07:00

test_batch_runner_durability.py

refactor: dedupe fallback warning per model, drive pool-cleanup tests through real run()

2026-08-01 15:05:05 +05:30

test_batch_runner_exit_code.py

fix(batch_runner): propagate fatal and validation errors as non-zero exit codes

2026-08-16 01:50:20 -07:00

test_bitwarden_secrets.py

…

test_cli_manual_compress.py

…

test_cli_quiet_stdout_leak.py

fix(cli): -Q stdout carries only the final response — no tool diffs, spinner lines, or reasoning

2026-08-23 17:01:31 -07:00

test_cli_skin_integration.py

…

test_code_skew.py

fix(dashboard): detect stale code after hermes update and refuse model picker with clear 503 (#86207)

2026-08-23 06:36:57 -07:00

test_command_secret_source.py

fix(gateway): hydrate cold profile secret sources

2026-08-02 00:11:50 -07:00

test_compression_watermark_commit.py

fix(compression): bound the rotation tail clone below the rotator's own flush

2026-08-15 23:37:22 -07:00

test_conftest_wal_gate.py

…

test_copilot_initiator.py

…

test_credential_file_permissions.py

…

test_cron_manage_profile_scope.py

fix(bot-mode): show profile-owned cronjobs

2026-08-17 16:34:02 -07:00

test_ctx_halving_fix.py

…

test_delegate_cascade_49148.py

…

test_desktop_update_shim_progress.py

test: shim-progress fake hermes answers the --keep-stash --help probe

2026-08-20 04:54:14 -07:00

test_desktop_update_windows_pipe_drain.py

fix(update): Windows Desktop updates finish instead of parking on "Updating Hermes" (#90937)

2026-08-20 12:41:17 -05:00

test_desktop_update_windows_progress.py

test(update): Windows progress self-test survives transient /progress socket stalls

2026-08-20 19:56:28 -07:00

test_desktop_update_windows_python_handoff.py

fix(update): Windows Desktop updates finish instead of parking on "Updating Hermes" (#90937)

2026-08-20 12:41:17 -05:00

test_desktop_update_windows_timestamp.py

fix(windows): avoid locale-sensitive update timestamps

2026-08-14 15:41:38 -05:00

test_dispatch_session_id.py

…

test_empty_model_fallback.py

feat(models): add qwen3.8-max to Nous portal + OpenRouter catalogs, replacing qwen3.7-max

2026-08-03 17:19:49 -07:00

test_empty_session_hygiene.py

…

test_engines_satisfiable.py

fix(install): provision Node 26 so managed npm satisfies engines

2026-08-14 21:48:14 -07:00

test_env_loader_applied_homes.py

…

test_env_loader_op_bootstrap.py

…

test_env_loader_secret_sources.py

fix(env): seed .op.env bootstrap into cold-profile hydration

2026-08-02 00:11:50 -07:00

test_estop.py

fix: post-merge audit follow-ups for #81138/#81139/#81141/#81148

2026-08-08 05:21:09 -07:00

test_evidence_store.py

…

test_fast_safe_load.py

…

test_fts_cjk_bigram.py

…

test_fts_update_of_narrowing.py

fix(state): quarantine CJK when ensure soft-fails after OF migration

2026-08-03 20:30:32 +05:30

test_gateway_streaming_nested_config.py

…

test_get_tool_definitions_cache_isolation.py

fix(plugins): isolate ownership by profile

2026-08-12 19:13:32 -07:00

test_gitlock.py

test(gitlock): pin the git-process guard in sweep tests (deflake slice 8)

2026-08-16 06:29:53 -07:00

test_hermes_bootstrap.py

test: run os-specific tests on their real host, not a faked one

2026-08-09 22:09:49 -04:00

test_hermes_constants.py

fix: widen secure_parent_dir to skip entire install tree

2026-08-24 15:42:09 +05:30

test_hermes_home_profile_warning.py

…

test_hermes_logging.py

test: run os-specific tests on their real host, not a faked one

2026-08-09 22:09:49 -04:00

test_hermes_state_compression_busy_retry.py

fix(compression): watermark commit — appends flow freely, concurrent tail survives compaction

2026-08-15 23:37:22 -07:00

test_hermes_state_compression_locks.py

test: run os-specific tests on their real host, not a faked one

2026-08-09 22:09:49 -04:00

test_hermes_state_readonly_preflight.py

…

test_hermes_state_wal_fallback.py

fix(state): reconcile salvaged WAL fixes with current main

2026-07-29 18:13:09 -07:00

test_hermes_state.py

fix(state): stop rebuilding the whole FTS index on every open when the trigram tokenizer is missing

2026-08-23 19:31:35 -07:00

test_hermetic_side_effect_guards.py

test: guard browser and Keychain side effects suite-wide (#35404)

2026-07-29 18:55:10 -07:00

test_honcho_client_concurrency.py

…

test_honcho_client_config.py

fix(honcho): honor explicit top-level apiKey on local base_urls; warn on keyless profile host blocks

2026-08-13 23:43:15 +05:30

test_honcho_session_context.py

…

test_honcho_startup_fail_open.py

fix(honcho): honor writeFrequency in sync_turn by routing through manager.save()

2026-08-13 23:43:15 +05:30

test_install_autostash_conflict_recovery.py

…

test_install_commit_pin_rollback.py

fix(installer): never let a stale --commit pin roll an install backwards

2026-07-29 17:35:16 -05:00

test_install_diverged_update.py

…

test_install_lockfile_churn.py

…

test_install_macos_launcher.py

…

test_install_no_initial_commit.py

…

test_install_ps1_ascii_only.py

…

test_install_ps1_browser_install.py

fix(install): stop npm-installing agent-browser eagerly in install.sh/install.ps1

2026-08-13 02:38:28 -07:00

test_install_ps1_managed_node_swap.py

test: pin Test-Node managed-Node swap to same-directory renames

2026-08-14 21:35:30 -07:00

test_install_ps1_native_stderr_eap.py

…

test_install_ps1_node_path_for_npm.py

…

test_install_ps1_python_fallback_venv.py

…

test_install_ps1_resolver_strictmode.py

fix(install.ps1): initialize LastResolver before the resolved-path report

2026-08-23 18:27:21 -07:00

test_install_ps1_uv_install_fallback.py

fix(install.ps1): surface uv installer errors and add GitHub + existing-uv fallbacks (#69216)

2026-08-14 22:33:58 -07:00

test_install_ps1_uv_powershell_host.py

…

test_install_ps1_venv_process_tree.py

fix(install): stop managed runtime child trees on Windows

2026-08-08 19:46:35 -07:00

test_install_ps1_venv_recreate_safety.py

test(installer): reproduce destructive venv fallback (#83149)

2026-08-14 21:58:09 -07:00

test_install_ps1_venv_rename_abort.py

test(install): lock no in-place venv gut on rename failure

2026-08-14 21:58:09 -07:00

test_install_ps1_venv_transaction_boundary.py

fix(installer): hold the venv rollback source through dependency validation (#83149)

2026-08-14 21:58:09 -07:00

test_install_ps1_web_server_syntax_probe.py

…

test_install_scripts_computer_use.py

fix(computer-use): verify Windows driver repair

2026-08-16 11:34:40 -07:00

test_install_sh_acp_launcher.py

fix(install): expose hermes-agent and hermes-acp launchers on PATH

2026-07-31 22:39:11 -07:00

test_install_sh_bootstrap_marker.py

…

test_install_sh_browser_install.py

fix(install): stop npm-installing agent-browser eagerly in install.sh/install.ps1

2026-08-13 02:38:28 -07:00

test_install_sh_install_method_stamp.py

…

test_install_sh_node_deps_failure.py

fix(install): a failed Node dependency install now fails the install instead of printing success (#85537)

2026-08-13 18:39:54 +00:00

test_install_sh_node_global_prefix.py

…

test_install_sh_node_npm_check.py

fix(install): a failed Node dependency install now fails the install instead of printing success (#85537)

2026-08-13 18:39:54 +00:00

test_install_sh_pythonpath_sanitization.py

…

test_install_sh_root_fhs_uv_python_path.py

…

test_install_sh_setup_wizard_tty_probe.py

…

test_install_sh_symlink_stomp.py

…

test_install_sh_termux_network_prereqs.py

…

test_install_unmerged_index.py

test(install): lock no in-place venv gut on rename failure

2026-08-14 21:58:09 -07:00

test_ipv4_preference.py

…

test_iron_proxy_cli.py

…

test_iron_proxy_e2e.py

…

test_iron_proxy.py

…

test_journal_mode_config.py

fix(state): make journal mode canonical and behaviorally verified

2026-07-29 18:13:09 -07:00

test_lazy_secrets_dispatch.py

fix(test): make lazy-secrets update-check E2E tests network-independent

2026-08-17 02:40:59 -07:00

test_lazy_secrets_import.py

test(lazy-secrets): fix CI compatibility — run from repo root, avoid live-system guard

2026-08-15 01:55:22 -07:00

test_lazy_session_regressions.py

…

test_list_recent_user_messages_handoffs.py

refactor(agent): fold simplify findings — DB picker parity, single scan, canonical strip delegation

2026-08-07 19:44:35 +05:30

test_live_system_guard_self_test.py

…

test_live_system_guard.py

test(tui_gateway): make the tui gateway suite order-independent

2026-07-29 21:30:53 -07:00

test_log_isolation.py

fix(tests): stop the suite writing into the operator's real Hermes logs

2026-07-29 19:53:17 -07:00

test_managed_runtime_resolution.py

fix: managed-runtime guard no longer trips on sdist/build copies in the workspace

2026-08-23 19:15:28 -07:00

test_mcp_serve.py

feat(mcp): migrate to the mcp 2.x SDK

2026-08-16 23:26:10 -07:00

test_message_reactions.py

fix: stamp _db_persisted at row load time so resumed transcripts never re-append (#92231)

2026-08-23 13:34:32 +05:30

test_mini_swe_runner.py

…

test_minimax_model_validation.py

…

test_minimax_oauth.py

fix(minimax-oauth): read streamed error bodies inside the client context + real-transport tests

2026-08-03 17:38:13 +05:30

test_minisweagent_path.py

…

test_moa_prepared_request_leak_78382.py

fix(agent): strip _moa_prepared_request before dispatching to native client

2026-08-14 21:36:41 -07:00

test_model_forces_max_completion_tokens.py

…

test_model_picker_scroll.py

…

test_model_tools_async_bridge.py

…

test_model_tools.py

Merge updated model metrics into tool metrics

2026-07-31 07:18:02 -07:00

test_no_shadowed_test_definitions.py

test: restore four silently shadowed definitions and guard against more

2026-07-31 22:35:16 -07:00

test_ollama_num_ctx.py

…

test_onepassword_secrets.py

…

test_os_marker_gating.py

test: convert the last host-OS fakes and guard double markers

2026-08-09 22:09:49 -04:00

test_output_cap_parsing.py

fix(model-metadata): parse 'exceeds model maximum output tokens' cap errors

2026-08-20 11:37:01 +05:30

test_packaging_build_guard.py

fix(packaging): ship bundled plugin manifests

2026-08-14 21:45:08 -07:00

test_packaging_metadata.py

fix(deps): exempt no-upload-date and exact-pinned packages from exclude-newer bricking

2026-08-14 21:50:48 -07:00

test_plugin_skills.py

fix: support packaged memory provider skills

2026-08-13 11:49:14 -07:00

test_plugin_storage.py

feat(plugins): per-plugin durable data directory that survives plugin update and removal

2026-08-17 15:12:53 -05:00

test_plugin_utils.py

…

test_plugins_manage_profile_scope.py

fix(desktop): clean up Plugins settings — drop Example Plugin, hide bundled built-ins, add profile scoping

2026-08-17 16:39:47 -07:00

test_process_loop_event_loop_warning.py

…

test_profile_isolation_runtime.py

…

test_project_metadata.py

Revert "remove Vercel AI Gateway and Vercel Sandbox (#33067)"

2026-07-29 19:48:37 -07:00

test_pty_keepalive_ws.py

fix(dashboard): redraw TUI after PTY reattach

2026-08-14 13:51:26 -07:00

test_pty_session.py

fix(dashboard): redraw TUI after PTY reattach

2026-08-14 13:51:26 -07:00

test_redaction_registry.py

feat(plugins): per-source pattern attribution + explicit pre-screen rebuild proof

2026-08-12 18:55:14 -07:00

test_resource_limits.py

fix(cli): never reap serve processes owned by a valid backend.lock.json

2026-08-10 17:02:56 -07:00

test_retry_utils.py

…

test_run_tests_parallel_stdio.py

fix(test-runner): native Windows venv probe + glyph-safe stdio

2026-07-29 21:30:53 -07:00

test_run_tests_parallel.py

fix(testing): tolerate legacy console encodings

2026-08-08 12:33:19 -07:00

test_sanitize_tool_error.py

…

test_schema_read_probe.py

fix(dashboard): derive the stale-schema read probe from SCHEMA_SQL

2026-08-07 00:41:58 -04:00

test_search_slow_query_log.py

…

test_secret_scope_plugin_families.py

fix(secrets): resolve google_meet realtime key via secret scope at spawn time

2026-08-02 10:04:48 -07:00

test_session_db_context_manager.py

docs(state): soften stale SessionDB self-pin wording after #88063

2026-08-16 20:43:13 -07:00

test_session_db_read_conn_pool.py

state: bound PEAK read connections with a permit, not just pooled returns

2026-08-10 17:02:56 -07:00

test_session_db_read_path_split.py

state: pool SessionDB read connections instead of leaking one per (SessionDB x thread)

2026-08-10 17:02:56 -07:00

test_session_skill_previews.py

fix(clients): hide compaction carriers across surfaces

2026-08-21 21:53:11 +05:30

test_session_system_prompt_dedup.py

fix(state): deduplicate session system prompts

2026-08-03 20:37:17 +05:30

test_session_vacuum_config.py

fix(state): make VACUUM interval configurable

2026-08-02 21:32:13 +05:30

test_session_workspace_binding.py

…

test_slack_thread_require_mention.py

…

test_slash_worker_watchdog.py

…

test_sql_injection.py

…

test_sqlite_lock_safe_inspection.py

fix(cli): untrack sqlite connections only after close succeeds

2026-07-31 23:18:12 -07:00

test_sqlite_wal_reset_gate.py

fix(db): never downgrade journal mode on a database with concurrent openers

2026-08-06 07:49:42 -07:00

test_stale_tool_call_marker_session_repair.py

fix(cli): back up state.db before clean-markers writes by default

2026-08-04 11:26:15 +05:30

test_stale_utils_module_import.py

…

test_state_db_malformed_repair.py

fix(state): fail closed on unscoped SQLite corruption

2026-08-23 02:55:03 +05:30

test_state_db_notadb_fail_closed.py

fix(state): fail closed on unscoped SQLite corruption

2026-08-23 02:55:03 +05:30

test_state_db_repair_live_writer_guard.py

fix(state): scope salvage to repair-connection durability + live-writer guard

2026-08-22 04:02:29 +05:30

test_state_db_repair_loop_cap.py

fix(state): bound the state.db repair loop and stop dead-backup accumulation (#86747)

2026-08-15 02:22:58 -07:00

test_state_db_repair_loop_mtime.py

fix(state): split forensic-backup identity from repair-epoch fingerprint; publish backup bundle atomically

2026-08-22 14:33:20 +05:30

test_state_db_repair_non_destructive.py

fix(state): make automatic repair non-destructive

2026-08-23 18:25:12 -07:00

test_state_db_stats.py

fix(state): recover FTS after orphan holder deferrals

2026-08-23 20:01:41 -07:00

test_state_db_write_durability.py

fix(state): make automatic repair non-destructive

2026-08-23 18:25:12 -07:00

test_subprocess_home_isolation.py

…

test_telegram_polling_progress_ptb.py

…

test_termux_all_extra_compat.py

…

test_timezone.py

…

test_tini_shim.py

…

test_toolset_distributions.py

…

test_toolsets.py

perf(toolsets): memoise resolve_toolset keyed on registry generation

2026-08-15 00:36:03 -07:00

test_trajectory_compressor_async.py

…

test_trajectory_compressor.py

…

test_transform_api_error_classification_hook.py

chore: trim hook test suite to core coverage and condense hooks docs

2026-08-13 09:36:02 -07:00

test_transform_llm_output_hook.py

…

test_transform_tool_result_hook.py

…

test_tui_entry_mcp_owner.py

fix(review): harden portable plugin boundaries

2026-08-07 09:44:21 -07:00

test_tui_gateway_loop_noise.py

…

test_tui_gateway_queue_on_busy.py

fix(tui): stop queued dispatch after session close

2026-08-16 06:26:37 -07:00

test_tui_gateway_server_crash_history.py

fix: preserve session history when a turn crashes

2026-08-06 23:08:23 -05:00

test_tui_gateway_server.py

fix(tui-gateway): revalidate transport ownership before sentinel-parking on WS disconnect

2026-08-23 18:58:40 -07:00

test_tui_gateway_ws.py

…

test_tui_mcp_late_refresh.py

fix(agent): the desktop's tools reach it on remote and cloud backends too

2026-08-06 19:35:47 -06:00

test_utils_atomic_roundtrip_yaml_save.py

fix(personality): preserve config comments in TUI/gateway config writes

2026-08-09 13:43:14 -07:00

test_utils_truthy_values.py

…

test_voice_max_recording_seconds.py

…

test_wal_checkpoint_strategy.py

fix(ci): sync lazy_deps SDK pins + update WAL vacuum test contract

2026-08-12 20:02:07 -07:00

test_web_server_sessiondb_eventloop.py

fix(dashboard): preserve maintenance writes on read polling

2026-08-02 21:30:54 +05:30

test_web_server_status_topology_cache.py

fix(serve): cache /api/status profile-gateway topology scan

2026-08-02 23:10:28 +05:30

test_web_server.py

fix(dashboard): treat Ctrl+C serve shutdown as a clean exit (supersedes #52970) (#85711)

2026-08-14 12:02:05 -04:00

test_windows_subprocess_no_window_flags.py

test(browser): cover warm_agent_browser_npx_cache's hardened behavior

2026-08-13 02:38:28 -07:00

test_ws_keepalive_config.py

fix(gateway): make ws keepalive and orphan-reap grace config-driven (#79635)

2026-08-23 17:43:39 -07:00

test_yaml_indent_consistency_31999.py

…

test_yuanbao_integration.py

…

test_yuanbao_markdown.py

…

test_yuanbao_pipeline.py

…

test_yuanbao_proto.py

…

test_yuanbao_reconnect_set_active.py

…

test_yuanbao_shutdown.py

…

test_zeroed_state_db.py

…