Patrick Buckley 476cce2e58 fix(streaming): scope stream bookkeeping to the send and discard dead segments server-side ...

Third review round on the retry window: two mediums fixed, one
observability gap closed.

- New UI-protocol method on_stream_discarded(): on_turn_committed clears
  only the inflight buffers — it cannot clear _ws_turn_content, the
  multi-segment buffer the IDLE payload drains, because earlier segments
  of a tool-looping turn must survive commits — so a dead attempt's text
  concatenated with the retried text in the dashboard's idle payload.
  SessionUIBase now truncates the turn buffer to a segment watermark
  (snapshotted in on_thinking_start, which precedes every stream
  segment), drops the never-displayed pending batch, and resets the
  inflight snapshot; the retry arm emits it in place of
  on_turn_committed. Server-side only — no SSE event, no client change;
  no-op on the CLI and eval UIs. Pinned with a real-SessionUIBase-buffer
  test: the recording fakes structurally cannot see this buffer.
- _active_stream_provider and _active_wire_msgs are send-scoped: cleared
  in send()'s finally, after the except arms' fatal formatting (the one
  legitimate fatal-path reader of the provider field). A later fatal on
  a utility lane falls back to self._provider instead of wearing a stale
  interactive-turn binding, and the full-context-sized wire fold no
  longer outlives its calibration use.
- stream.retry carries dead_usage and dead_content_chars: the abandoned
  generation's billed tokens are otherwise invisible (the wire reports
  usage only at stream end — Anthropic's early prompt tokens arrive, the
  OpenAI chat lane's usage chunk trails the finish), so the log line
  records what the wire delivered plus the discarded completion's char
  count for spend reconciliation.

2026-08-04 04:53:17 -07:00

..

data/wire_payloads

fix(model-turn): one sampling-knob assignment scheme — alias > config > model definition > omit

2026-07-13 08:48:27 -07:00

__init__.py

Initial commit — turnstone multi-node AI orchestration platform.

2026-03-02 00:33:37 -08:00

_coord_test_helpers.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

_helpers.py

fix(watch): harden nudge/wake delivery across eviction, cancel, and identity rebinds

2026-07-07 16:42:36 -07:00

_js_harness_helpers.py

feat(ui): copy-to-clipboard for messages and rendered blocks

2026-08-02 06:15:40 -07:00

_oidc_test_helpers.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

_proc_helpers.py

feat(bash): opt-in background shells with delta output reader and kill tool (#817)

2026-07-10 15:42:33 -07:00

_replay_helpers.py

test(approvals): concurrency matrix + suite migration to the cycle model

2026-07-05 01:57:54 -07:00

_session_helpers.py

fix(providers): PR review — orphan deltas arm the finish shim, test style

2026-07-13 22:39:19 -07:00

_sse_recovery_helpers.py

test(sse): end-to-end recovery harness (server-contract + browser livepass)

2026-07-20 22:38:32 -07:00

_sse_recovery_server.py

test(e2e): E8 observes the double render; honest non-vacuity for E6 (#900 r2)

2026-07-24 18:15:03 -07:00

_wire_capture.py

test(providers): ladder-to-wire effort parity harness across all lanes

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

conftest.py

feat(server): opt-in server.require_project gate

2026-07-18 16:12:59 -07:00

spike_sdk_concurrency.py

test(mcp): SDK 1.27 concurrency spike for per-(user, server) pool

2026-05-05 15:27:14 -07:00

test_admin_calibrate_endpoint.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_admin_create_admin.py

fix(admin): add create-admin CLI; stop run.sh onboarding into a role-less user

2026-07-11 18:15:48 -07:00

test_admin_db_config.py

feat(admin): align turnstone-admin DB config with server (config.toml + env) (#531)

2026-05-19 08:09:33 -07:00

test_admin_export.py

feat(export): export workstream conversations as OpenAI messages JSON

2026-05-29 21:11:16 -07:00

test_admin_model_registry_refresh.py

refactor: single-style module imports and narrowed JSON body typing

2026-08-03 20:11:28 -07:00

test_admin_rerank_calibrate.py

feat(rerank): per-model calibration on detect + calibrate endpoint

2026-06-01 15:47:14 -07:00

test_api_versioning.py

chore(tests): refresh fixtures for path-keyed URL family

2026-04-26 22:14:22 -07:00

test_app_js.py

feat(ui): copy-to-clipboard for messages and rendered blocks

2026-08-02 06:15:40 -07:00

test_attachment_buffer.py

test: satisfy ruff-format and lift side-effects out of asserts

2026-06-04 11:03:13 -07:00

test_attachments_pdf_audio.py

fix(attachments): address fix-review nits (ftyp scan, text-preview, cache doc)

2026-06-16 00:48:14 -07:00

test_audio.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_audit.py

refactor(coordinator): rename task_list → tasks, doc/prompt sweep (#437)

2026-04-27 22:51:31 -07:00

test_auth_identity.py

fix(auth): isolate server/console session cookies by name

2026-06-15 02:48:50 -07:00

test_auth.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_background_shells.py

fix(bash): report exit code for killed shells; single import style in registry tests

2026-07-10 15:42:33 -07:00

test_bash_background_tool.py

fix(bash): report exit code for killed shells; single import style in registry tests

2026-07-10 15:42:33 -07:00

test_bash_tool_background_hang.py

feat(sse): coalesce tool_output_chunk emission per call_id

2026-07-20 22:38:32 -07:00

test_bm25.py

fix(tool-search): discovery-failure records are per-user, rerank counts honest

2026-08-02 01:48:30 -07:00

test_cancel.py

refactor(session): extract think-tag splitting into ThinkTagSplitter

2026-08-04 04:53:17 -07:00

test_canonicalize_skill_string_list.py

feat(skills): parse and store Anthropic spec paths + uplift columns

2026-05-23 14:20:36 -07:00

test_channel_discord.py

fix(channels): suppress mention resolution and escape untrusted fields

2026-07-29 22:11:38 -07:00

test_channel_protocol.py

refactor(channels): backfill review of Slack/Discord adapters (#382)

2026-04-18 05:49:54 -07:00

test_channel_routing.py

test(approvals): concurrency matrix + suite migration to the cycle model

2026-07-05 01:57:54 -07:00

test_channel_slack.py

fix(channels): suppress mention resolution and escape untrusted fields

2026-07-29 22:11:38 -07:00

test_channel_sse.py

refactor(channels): backfill review of Slack/Discord adapters (#382)

2026-04-18 05:49:54 -07:00

test_channel_storage.py

feat: add PostgreSQL CI integration tests (#156)

2026-03-23 11:11:40 -07:00

test_chat_session_interactivity_flag.py

feat(mcp): admin status, deferred-consent persistence, operator docs (Phase 9) (#516)

2026-05-12 13:15:09 -07:00

test_child_event_bus.py

feat(console): event-driven wait_for_workstream + idle cleanup via ChildEventBus

2026-05-11 01:06:47 -07:00

test_child_source.py

feat(coord): Stage 3 SessionManager Children primitive lift + cluster bus push paths

2026-04-30 23:20:14 -07:00

test_children_registry.py

feat(coord): Stage 3 SessionManager Children primitive lift + cluster bus push paths

2026-04-30 23:20:14 -07:00

test_close_reason_persistence.py

rbac: builtin-role override editor + tighten under-enforced perm gates (#585)

2026-05-24 18:31:23 -07:00

test_collector_reachability.py

fix(tls): repair cluster mTLS — cert identity, renewal scoping, hot-reload

2026-05-30 16:27:15 -07:00

test_compaction_checkpoint.py

feat(compaction): lifecycle events, web progress card, history re-render

2026-07-17 11:28:44 -07:00

test_compaction_crossing.py

feat(coordinator): carry task and child handles across a compaction

2026-07-29 22:11:38 -07:00

test_config_store.py

refactor(tools): remove man, math, and plan_agent built-in tools

2026-05-31 19:54:43 -07:00

test_config.py

fix(tools): lower working directory and workspace into fs tool descriptions

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

test_console_available_models.py

fix(providers): the session effort level always reaches the local-lane wire

2026-07-05 01:59:38 -07:00

test_console_effort_ladder.py

fix(providers): the session effort level always reaches the local-lane wire

2026-07-05 01:59:38 -07:00

test_console_idle_cleanup.py

feat(console): event-driven wait_for_workstream + idle cleanup via ChildEventBus

2026-05-11 01:06:47 -07:00

test_console_metrics.py

refactor(core): unify approve_tools across both kinds (#436)

2026-04-27 21:52:57 -07:00

test_console_route_attachments.py

feat: workstream attachments at creation time + SDK + UI parity (#362)

2026-04-16 13:30:25 -07:00

test_console_router.py

refactor(routing): replace hash-ring rebalancer with rendezvous (HRW)… (#384)

2026-04-18 19:02:52 -07:00

test_console_routing_proxy.py

test(personas): harden the guard suite — real paths over scripted events

2026-07-03 00:29:26 -07:00

test_console_session_factory.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_console.py

fix(console): restore back-to-console from a proxied node view

2026-07-25 00:50:29 -07:00

test_conversation_css.py

fix(ui): interactive tool cards no longer collapse to an empty stripe

2026-06-08 10:08:30 -07:00

test_conversation_js.py

fix(compaction): validate retry_in backoff before rendering the retry note

2026-07-17 11:28:44 -07:00

test_cooperative_compaction.py

fix(session): survive mid-stream transport deaths in interactive turns (#937)

2026-08-04 04:53:17 -07:00

test_coord_rich_ws_state_payload.py

chore(compaction): review round 10 — cleanups from the first correctness-clean round

2026-07-17 11:28:44 -07:00

test_coord_ui_approve_tools.py

test(approvals): concurrency matrix + suite migration to the cycle model

2026-07-05 01:57:54 -07:00

test_coordinator_adapter.py

fix(watch): harden nudge/wake delivery across eviction, cancel, and identity rebinds

2026-07-07 16:42:36 -07:00

test_coordinator_client.py

feat(coordinator): nudge a coordinator that goes idle holding unfinished work

2026-07-29 22:11:38 -07:00

test_coordinator_close_all_children.py

feat(cancel): honest cancellation dispositions + coordinator subtree propagation

2026-06-26 03:28:06 -07:00

test_coordinator_end_to_end.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_coordinator_endpoints.py

fix: address review — request-scoped storage in coord tenancy checks

2026-07-06 19:16:09 -07:00

test_coordinator_governance.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_coordinator_idle_observer.py

fix(coordinator): the interjection handoff cannot lose the message, and the fact block is bounded

2026-07-29 22:11:38 -07:00

test_coordinator_page.py

feat(coordinator): nudge a coordinator that goes idle holding unfinished work

2026-07-29 22:11:38 -07:00

test_coordinator_proxy_auth.py

feat(coordinator): coordinator workstream kind — phase 1 (#368)

2026-04-16 21:40:50 -07:00

test_coordinator_tools.py

feat(coordinator): nudge a coordinator that goes idle holding unfinished work

2026-07-29 22:11:38 -07:00

test_copy_actions_js.py

fix(ui): block-copy dismissal listens on documentElement, not document

2026-08-02 06:15:40 -07:00

test_cwd_tool_notes.py

docs(tools): apply review round-1 findings (cwd notes)

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

test_db.py

feat: structured memory system — typed/scoped memories with BM25 rele… (#53)

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

test_deadline.py

fix(providers): PR review — orphan deltas arm the finish shim, test style

2026-07-13 22:39:19 -07:00

test_docker_healthcheck.py

test: address Copilot review on the leaked-thread guard

2026-06-17 18:07:45 -07:00

test_doctor.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_drain_stream.py

refactor(streaming): dedupe transport conversion, usage merge, cancel finalize

2026-08-04 04:53:17 -07:00

test_edit_file.py

Fix/codeql quality findings (#299)

2026-04-04 17:53:20 -07:00

test_edit.py

Initial commit — turnstone multi-node AI orchestration platform.

2026-03-02 00:33:37 -08:00

test_effort_ladder_wire_parity.py

fix(providers): the session effort level always reaches the local-lane wire

2026-07-05 01:59:38 -07:00

test_effort_ladder.py

feat(providers): drop o-series and pre-5.4 GPT-5 capability rows

2026-07-13 22:39:19 -07:00

test_env_scrub.py

Bound search tool output against pathological inputs (#473)

2026-05-04 00:46:43 -07:00

test_eval_core.py

feat(eval): behavioral eval for the coordinator idle nudges

2026-07-29 22:11:38 -07:00

test_eval_nudges.py

fix(coordinator): the interjection handoff cannot lose the message, and the fact block is bounded

2026-07-29 22:11:38 -07:00

test_export.py

refactor(storage): drop dead ws_id/user_id from workstream_attachments

2026-06-04 11:03:13 -07:00

test_fence.py

fix(fence): guard detection_pattern against an empty tag set

2026-06-26 19:37:01 -07:00

test_fts5.py

Add pluggable storage backend (SQLite + PostgreSQL) and deployment packaging (#20)

2026-03-03 22:57:34 -08:00

test_global_sse_boot_epoch.py

docs(#881): faithful token_hex(8) test epoch; document inline shutdown put

2026-07-22 23:44:04 -07:00

test_governance_endpoints.py

fix(personas): close review findings across the envelope, resume, and RBAC lanes

2026-07-03 00:29:26 -07:00

test_governance_storage.py

test(personas): harden the guard suite — real paths over scripted events

2026-07-03 00:29:26 -07:00

test_hatch_js.py

fix: CI lint pin + copilot-thread hardening

2026-06-10 13:31:53 -07:00

test_health_tls_state.py

fix(tls): mTLS-aware container healthcheck + boot-time init retry

2026-06-09 18:37:46 -07:00

test_healthcheck.py

feat: multi-model health tracking with runtime default and DB-only st… (#304)

2026-04-04 23:44:29 -07:00

test_history_decoration.py

fix(model-turn): round-3 review — complete the scheme rollout to the main loop, coordinator role, admin save path, and CLI switch

2026-07-13 08:48:27 -07:00

test_history_projection.py

fix(ui): align system-turn row event_id with its SSE event (no double-render)

2026-06-04 11:03:13 -07:00

test_html.py

fix(web): preserve block structure in strip_html, remove ReDoS risk

2026-05-30 01:11:48 -07:00

test_idle_nudge_wake_integration.py

fix(coordinator): the interjection handoff cannot lose the message, and the fact block is bounded

2026-07-29 22:11:38 -07:00

test_idle_nudge_watcher.py

fix(compaction): review round 8 — seam obligations become primitives

2026-07-17 11:28:44 -07:00

test_images.py

fix(attachments): normalize EXIF orientation so thumbnails and models see upright images

2026-06-16 03:41:51 -07:00

test_interactive_adapter.py

refactor(tools): remove man, math, and plan_agent built-in tools

2026-05-31 19:54:43 -07:00

test_interactive_pane_js.py

fix(#900): close two vacuity holes the round-2 scenarios left open

2026-07-24 18:15:03 -07:00

test_is_error_persistence.py

feat(storage): persist tool-result is_error (migration 060)

2026-06-04 11:03:13 -07:00

test_judge_heuristic.py

refactor(tools): remove man, math, and plan_agent built-in tools

2026-05-31 19:54:43 -07:00

test_judge_rules_storage.py

Worktree feat configurable output guard (#305)

2026-04-05 01:53:33 -07:00

test_judge_storage.py

fix(storage): per-row conflict tolerance on bulk verdict insert

2026-06-09 20:00:55 -07:00

test_judge.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_logging.py

Add structured logging with structlog and context propagation

2026-03-04 06:47:30 -08:00

test_lowering.py

feat(task-agent): carry the provider-native reasoning lane in the sub-harness

2026-07-11 16:37:13 -07:00

test_markdown.py

Quality overhaul: code tooling, CI/CD, architecture diagrams, UI rede… (#1)

2026-03-02 16:55:12 -08:00

test_mcp_admin_api.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_mcp_admin_bulk_revoke.py

fix(mcp): complete oauth_obo revocation lifecycle + fix hot-path regression (follow-up review)

2026-07-12 19:03:35 -07:00

test_mcp_client.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_mcp_consent_url_sibling_audit.py

fix(mcp): round-3 review fixes — revocation generation for catalog publishers

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

test_mcp_crypto.py

feat(mcp): per-(user, server) OAuth 2.1 + PKCE flow

2026-05-04 22:00:23 -07:00

test_mcp_hot_reload.py

fix(tool-search): discovery-failure records are per-user, rerank counts honest

2026-08-02 01:48:30 -07:00

test_mcp_http_parsers.py

feat(mcp): SDK 401/403 introspection via httpx response hook

2026-05-06 11:29:25 -07:00

test_mcp_integration.py

refactor(mcp): consolidate per-server state into StaticServerState dataclass

2026-05-04 22:00:23 -07:00

test_mcp_live_flaky_server.py

fix(mcp): review round 4 — removal/reconcile lifecycle, honest skip reporting, same-kind debounce recovery

2026-07-14 11:39:25 -07:00

test_mcp_live_push_refresh.py

fix(mcp): review round 4 — removal/reconcile lifecycle, honest skip reporting, same-kind debounce recovery

2026-07-14 11:39:25 -07:00

test_mcp_oauth_connections.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_mcp_oauth_discovery.py

fix(mcp): harden Entra OBO OAuth review follow-ups for #706

2026-06-25 20:42:22 -07:00

test_mcp_oauth_handlers.py

Drop dead last_ws_id/last_tool_call_id columns from mcp_pending_consent

2026-07-04 03:27:55 -07:00

test_mcp_oauth_pending_storage.py

feat(mcp): oauth schema + minimum admin form

2026-05-04 22:00:23 -07:00

test_mcp_oauth_pkce.py

feat(mcp): per-(user, server) OAuth 2.1 + PKCE flow

2026-05-04 22:00:23 -07:00

test_mcp_oauth_refresh.py

fix(oidc/mcp): make runtime OIDC rediscovery actually work; preserve oauth_user paths

2026-07-12 19:03:35 -07:00

test_mcp_oauth_revoke.py

feat(mcp): per-user MCP server consent UX (Phase 8)

2026-05-07 13:59:50 -07:00

test_mcp_oauth_storage.py

feat(mcp): per-user MCP server consent UX (Phase 8)

2026-05-07 13:59:50 -07:00

test_mcp_obo_mint.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_mcp_pending_consent_dispatch.py

feat(mcp): admin status, deferred-consent persistence, operator docs (Phase 9) (#516)

2026-05-12 13:15:09 -07:00

test_mcp_pending_consent_endpoints.py

Drop dead last_ws_id/last_tool_call_id columns from mcp_pending_consent

2026-07-04 03:27:55 -07:00

test_mcp_pending_consent_storage.py

fix(mcp): complete oauth_obo revocation lifecycle + fix hot-path regression (follow-up review)

2026-07-12 19:03:35 -07:00

test_mcp_pool_auth_integration.py

test: eliminate leaked-thread test pollution + guard against it

2026-06-17 18:07:45 -07:00

test_mcp_pool_auth_introspection.py

fix(mcp): review round 1 — supersede retired-lock refreshes, per-kind debounce, complete gather pairs

2026-07-14 11:39:25 -07:00

test_mcp_pool_auth_prompt_integration.py

test: eliminate leaked-thread test pollution + guard against it

2026-06-17 18:07:45 -07:00

test_mcp_pool_auth_resource_integration.py

test: eliminate leaked-thread test pollution + guard against it

2026-06-17 18:07:45 -07:00

test_mcp_pool_owner.py

fix(mcp): round-2 review fixes for the catalog-retention branch

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

test_mcp_prompt_sync.py

feat: wire prompt templates into session startup with full creation-p… (#47)

2026-03-12 16:57:31 -07:00

test_mcp_registry_api.py

fix: align ConfigStore implementation with spec (#153)

2026-03-21 16:12:50 -07:00

test_mcp_registry.py

fix: validate URL scheme after MCP registry template substitution (#133)

2026-03-20 20:01:26 -07:00

test_mcp_server_storage.py

feat: add PostgreSQL CI integration tests (#156)

2026-03-23 11:11:40 -07:00

test_mcp_token_store_metadata.py

feat(mcp): per-user MCP server consent UX (Phase 8)

2026-05-07 13:59:50 -07:00

test_mcp_token_store.py

feat(mcp): per-(user, server) OAuth 2.1 + PKCE flow

2026-05-04 22:00:23 -07:00

test_mcp_transport_owner.py

fix(mcp): route pool transport lifecycles through per-entry owner tasks (#788)

2026-07-06 18:46:28 -07:00

test_mcp_user_catalog.py

fix(mcp): review round 1 — supersede retired-lock refreshes, per-kind debounce, complete gather pairs

2026-07-14 11:39:25 -07:00

test_mcp_user_pool.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_memory_api.py

feat(projects): governed project containers — memory scope, grouping, manage UI (#724)

2026-06-26 17:24:06 -07:00

test_memory_relevance.py

feat(projects): governed project containers — memory scope, grouping, manage UI (#724)

2026-06-26 17:24:06 -07:00

test_metacognition.py

fix(coordinator): the interjection handoff cannot lose the message, and the fact block is bounded

2026-07-29 22:11:38 -07:00

test_midstream_retry.py

fix(streaming): scope stream bookkeeping to the send and discard dead segments server-side

2026-08-04 04:53:17 -07:00

test_migration_049.py

feat(mcp): oauth schema + minimum admin form

2026-05-04 22:00:23 -07:00

test_migration_060.py

fix: pre-push review — page migration 060 backfill, heal legacy orphan tool_use on load

2026-06-04 11:03:13 -07:00

test_migration_061.py

feat(memory): durable per-user coordinator scope + anonymous-coordinator guard

2026-06-12 13:54:49 -07:00

test_migration_062.py

feat(projects): governed project containers — memory scope, grouping, manage UI (#724)

2026-06-26 17:24:06 -07:00

test_migration_063.py

refactor(personas): file-backed built-in prompts, explicit source column

2026-07-03 00:29:26 -07:00

test_migration_065.py

Addendum to Entra ID's... proclivities (#772)

2026-07-04 16:48:52 -07:00

test_migration_066.py

feat(schedules): add persona and project settings to scheduled tasks

2026-07-08 00:59:14 -07:00

test_migration_concurrency.py

fix(storage): prevent concurrent-migration deadlock on the advisory lock

2026-05-31 16:05:21 -07:00

test_model_definition_storage.py

fix(reasoning): apply full-stack review findings

2026-05-09 02:45:13 -07:00

test_model_probe.py

feat(providers): drop o-series and pre-5.4 GPT-5 capability rows

2026-07-13 22:39:19 -07:00

test_model_provider_obo.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_model_registry.py

refactor: single-style module imports and narrowed JSON body typing

2026-08-03 20:11:28 -07:00

test_model_turn.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_models_changed_event.py

feat(providers): anthropic-compatible lane for local /v1/messages servers

2026-06-11 20:27:29 -07:00

test_native_tool_calls_mirror.py

fix: pre-push review — page migration 060 backfill, heal legacy orphan tool_use on load

2026-06-04 11:03:13 -07:00

test_node_info.py

feat(node): auto-detect node capabilities via kernel interfaces (#445)

2026-04-28 13:49:31 -07:00

test_node_metadata_storage.py

feat: add per-node metadata with auto-collection, admin API, and cons… (#318)

2026-04-06 03:08:19 -07:00

test_notify_completion.py

refactor(core): session.messages is the canonical Turn trajectory

2026-06-04 11:03:13 -07:00

test_notify_dispatcher.py

fix(notify): unbreak PG test backend on the notify dispatcher suite

2026-05-11 01:23:54 -07:00

test_notify_http.py

feat: deliver scheduled workstream results to Discord on completion (#308)

2026-04-05 17:18:26 -07:00

test_notify_tool.py

feat(compaction): lifecycle events, web progress card, history re-render

2026-07-17 11:28:44 -07:00

test_nudge_queue.py

feat(coordinator): idle nudges deliver only on the idle wake, and a queued interjection owns the seam

2026-07-29 22:11:38 -07:00

test_oauth_ssrf.py

fix(oidc): carry the opt-in hint on discovered-endpoint rejections

2026-07-06 21:52:42 -07:00

test_oidc_credential_storage.py

fix(oidc/mcp): login-path self-heal, guard mint persist, CAS credential rotation

2026-07-12 19:03:35 -07:00

test_oidc_handlers.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_oidc_storage.py

Addendum to Entra ID's... proclivities (#772)

2026-07-04 16:48:52 -07:00

test_oidc.py

fix(oidc): trust Entra's graph.microsoft.com userinfo out of the box

2026-07-12 19:03:35 -07:00

test_open_preview_tool.py

fix(preview): fetch ceiling tracks the widest kind cap, not a flat 10 MB

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

test_openapi.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_operator_instruction_declaration.py

fix(core): legalize malformed tool-call arguments before the wire

2026-07-05 11:59:52 -07:00

test_orphan_cleanup.py

fix(storage): enforce orphan-ness inside the purge DELETE + chunk IN-lists

2026-06-11 14:00:42 -07:00

test_output_assessment_storage.py

feat(judge): output_guard LLM stage with capability gate (#560 mitigation #1)

2026-05-24 17:49:27 -07:00

test_output_guard_judge.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_output_guard.py

fix(redaction): match connection-string schemes case-insensitively

2026-07-06 21:42:51 -07:00

test_pdf.py

fix(attachments): address PR review feedback (Copilot + code-quality)

2026-06-16 00:48:14 -07:00

test_per_user_message_context.py

fix(session): address ultrareview findings on shared-workstream branch

2026-07-02 19:23:04 -07:00

test_perception.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_persona_endpoints.py

refactor(personas): file-backed built-in prompts, explicit source column

2026-07-03 00:29:26 -07:00

test_persona_guards.py

feat(coordinator): nudge a coordinator that goes idle holding unfinished work

2026-07-29 22:11:38 -07:00

test_persona_snapshot.py

fix(personas): repr the input in the not-found resolver error

2026-07-06 19:58:36 -07:00

test_persona_storage.py

refactor(personas): file-backed built-in prompts, explicit source column

2026-07-03 00:29:26 -07:00

test_phase6_endpoints.py

fix: scope private-project workstream visibility to members, not admins

2026-07-06 19:16:09 -07:00

test_preview_js.py

fix(ui): ephemeral panes close on split-dismiss instead of orphaning a tab

2026-07-07 22:07:20 -07:00

test_preview.py

feat(preview): probe preflight, legacy charsets, remote-assets opt-in, md vendor parity

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

test_project_api.py

feat(projects): saved-list project column + per-project resources view

2026-07-01 18:01:50 -07:00

test_project_memory_scope.py

feat(projects): governed project containers — memory scope, grouping, manage UI (#724)

2026-06-26 17:24:06 -07:00

test_project_storage.py

fix(projects): close review-found tenancy leaks + correctness regressions

2026-07-01 18:01:50 -07:00

test_project_workstream_visibility.py

fix: scope private-project workstream visibility to members, not admins

2026-07-06 19:16:09 -07:00

test_prompt_templates_runtime.py

fix(skills): harden task_agent / persona / skill activation from whole-PR review

2026-07-03 19:16:36 -07:00

test_prompts.py

refactor(personas): file-backed built-in prompts, explicit source column

2026-07-03 00:29:26 -07:00

test_provider_anthropic_compat.py

feat(providers): one transport — drain create_streaming, retire create_completion (#831)

2026-07-13 22:39:19 -07:00

test_provider_anthropic_reasoning.py

fix(reasoning): apply full-stack review findings

2026-05-09 02:45:13 -07:00

test_provider_anthropic_replay.py

perf: dict-native wire-prep — drop the per-send Turn<->dict round-trip

2026-06-04 11:03:13 -07:00

test_provider_data_envelope.py

feat(storage): tag the native lane with its producer ({producer, blocks})

2026-06-04 11:03:13 -07:00

test_provider_openai_responses_reasoning.py

feat(task-agent): carry the provider-native reasoning lane in the sub-harness

2026-07-11 16:37:13 -07:00

test_provider_xai.py

feat(providers): one transport — drain create_streaming, retire create_completion (#831)

2026-07-13 22:39:19 -07:00

test_providers_document_parts.py

feat(providers): drop o-series and pre-5.4 GPT-5 capability rows

2026-07-13 22:39:19 -07:00

test_providers.py

feat(anthropic): onboard claude-opus-5

2026-07-25 01:12:04 -07:00

test_ratelimit.py

fix: apply /review feedback on legacy URL cleanup

2026-04-26 22:14:22 -07:00

test_reasoning_audit_log_discipline.py

fix(session): survive mid-stream transport deaths in interactive turns (#937)

2026-08-04 04:53:17 -07:00

test_recall_compaction_scope.py

feat(recall): scope the recall tool to the compacted past

2026-07-01 21:05:57 -07:00

test_reconstruct_messages.py

feat(operator-context): deliver structured per-kind meta to the UI

2026-06-04 11:03:13 -07:00

test_renderer_js.py

feat(ui): copy-to-clipboard for messages and rendered blocks

2026-08-02 06:15:40 -07:00

test_rendezvous.py

refactor(routing): replace hash-ring rebalancer with rendezvous (HRW)… (#384)

2026-04-18 19:02:52 -07:00

test_require_project.py

feat(coordinator): extend server.require_project to coordinator creates

2026-07-20 11:16:08 -07:00

test_rerank_calibrate.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_rerank.py

refactor(rerank): reranker is a per-model definition only (drop global endpoint settings)

2026-06-01 21:34:12 -07:00

test_resume_flow.py

feat: replace Redis MQ with direct HTTP transport (Phase 1)

2026-03-30 20:30:05 -07:00

test_rewind_retry.py

fix(#894): cold flights key on None; typed generation access; abort-Set producer pins

2026-07-24 15:04:31 -07:00

test_route_proxy_audit.py

refactor(tools): remove man, math, and plan_agent built-in tools

2026-05-31 19:54:43 -07:00

test_rule_registry.py

refactor(tools): remove man, math, and plan_agent built-in tools

2026-05-31 19:54:43 -07:00

test_safety.py

Initial commit — turnstone multi-node AI orchestration platform.

2026-03-02 00:33:37 -08:00

test_saved_handler_unified.py

test(personas): harden the guard suite — real paths over scripted events

2026-07-03 00:29:26 -07:00

test_schedule_api.py

feat(schedules): add persona and project settings to scheduled tasks

2026-07-08 00:59:14 -07:00

test_scheduled_tasks_storage.py

feat(schedules): add persona and project settings to scheduled tasks

2026-07-08 00:59:14 -07:00

test_scheduler.py

feat(schedules): add persona and project settings to scheduled tasks

2026-07-08 00:59:14 -07:00

test_schema_parity.py

refactor(personas): file-backed built-in prompts, explicit source column

2026-07-03 00:29:26 -07:00

test_scoring.py

Quality overhaul: code tooling, CI/CD, architecture diagrams, UI rede… (#1)

2026-03-02 16:55:12 -08:00

test_sdk_console.py

feat(schedules): add persona and project settings to scheduled tasks

2026-07-08 00:59:14 -07:00

test_sdk_events.py

refactor(tools): remove man, math, and plan_agent built-in tools

2026-05-31 19:54:43 -07:00

test_sdk_governance.py

fix: add admin.prompt_policies to valid permissions and builtin-admin role (#303)

2026-04-04 22:19:03 -07:00

test_sdk_server_attachments.py

feat: workstream attachments at creation time + SDK + UI parity (#362)

2026-04-16 13:30:25 -07:00

test_sdk_server.py

feat(schedules): add persona and project settings to scheduled tasks

2026-07-08 00:59:14 -07:00

test_sdk_sse.py

docs: mark 1.5.0 legacy URL surface removal

2026-04-26 22:14:22 -07:00

test_sdk_stream_boundary.py

fix(streaming): finalize dead attempts on terminal paths and harden the retry window

2026-08-04 04:53:17 -07:00

test_sdk_sync.py

refactor(core): lift list + saved verb bodies across both kinds (Stage 2 verb lift)

2026-04-26 13:11:07 -07:00

test_search_history_visibility.py

feat(recall): scope the recall tool to the compacted past

2026-07-01 21:05:57 -07:00

test_server_attachments_endpoints.py

chore(compaction): review round 10 — cleanups from the first correctness-clean round

2026-07-17 11:28:44 -07:00

test_server_attachments_on_create.py

fix(server): sanitize the retry closure's error display

2026-07-17 20:08:56 -07:00

test_server_authz.py

fix(session): force-cancel runs the abandon machinery before emitting idle

2026-07-29 22:11:38 -07:00

test_server_available_models.py

feat(audio): voice I/O — speech-to-text + text-to-speech via model roles (#618)

2026-05-30 14:03:39 -07:00

test_server_compat.py

feat(providers): share the effort-knob→chat_template_kwargs mapping with the openai-compatible lane

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

test_server_lifespan_mcp_crypto.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_server_live.py

feat(personas): core stamping, four-lever application, create/spawn/SDK threading; remove /creative; add --persona

2026-07-03 00:29:26 -07:00

test_server_node_models_metadata.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_service_auth_boundary.py

feat(console): forward Last-Event-ID through SSE proxy

2026-05-22 15:13:39 -07:00

test_services_storage.py

refactor: channel router + scheduler use SDK clients

2026-03-30 20:30:05 -07:00

test_session_attachments.py

feat(models): default-deny governance and admin UI for per-alias backend auth

2026-08-03 20:11:28 -07:00

test_session_backend_error_format.py

fix(streaming): close the retry window's generation, identity, and masking holes

2026-08-04 04:53:17 -07:00

test_session_chat_reasoning_replay.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_session_manager.py

feat(personas): core stamping, four-lever application, create/spawn/SDK threading; remove /creative; add --persona

2026-07-03 00:29:26 -07:00

test_session_mcp_dispatch_error.py

fix(effect-status): harden effect_status decode + fix tests for typed synth

2026-06-26 08:38:31 -07:00

test_session_replay_reasoning.py

feat(providers): one transport — drain create_streaming, retire create_completion (#831)

2026-07-13 22:39:19 -07:00

test_session_routes.py

feat(preview): rich preview pane + open_preview tool

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

test_session_synth_reasoning_block.py

fix(session): survive mid-stream transport deaths in interactive turns (#937)

2026-08-04 04:53:17 -07:00

test_session_ui_base.py

test(sse): normalize session_ui_base imports to a single style

2026-07-07 23:32:20 -07:00

test_session_worker.py

fix(compaction): review round 8 — seam obligations become primitives

2026-07-17 11:28:44 -07:00

test_session.py

fix(session): survive mid-stream transport deaths in interactive turns (#937)

2026-08-04 04:53:17 -07:00

test_sessions.py

refactor: single-style module imports and narrowed JSON body typing

2026-08-03 20:11:28 -07:00

test_settings_api.py

feat(web-search): replace Tavily/DuckDuckGo backends with self-hosted SearxNG

2026-05-31 19:03:16 -07:00

test_settings_registry.py

refactor(tools): remove man, math, and plan_agent built-in tools

2026-05-31 19:54:43 -07:00

test_shell_js.py

feat(ui): copy-to-clipboard for messages and rendered blocks

2026-08-02 06:15:40 -07:00

test_skill_adherence.py

fix(eval): address Copilot review — checkout-agnostic docs + skill validation

2026-07-03 17:30:33 -07:00

test_skill_discovery_api.py

chore(skills): drop Anthropic attribution from SKILL.md spec references

2026-05-24 14:56:43 -07:00

test_skill_parse_api.py

chore(skills): drop Anthropic attribution from SKILL.md spec references

2026-05-24 14:56:43 -07:00

test_skill_parser.py

chore(skills): drop Anthropic attribution from SKILL.md spec references

2026-05-24 14:56:43 -07:00

test_skill_resource_materialization.py

refactor(skills): move applied-skill body out of the identity system message

2026-07-03 19:16:36 -07:00

test_skill_resources_api.py

feat: skill resources — API, admin UI, runtime injection, and SDK

2026-03-17 01:26:44 -07:00

test_skill_resources_storage.py

feat: add PostgreSQL CI integration tests (#156)

2026-03-23 11:11:40 -07:00

test_skill_scanner.py

feat: skill scanner — evaluate SKILL.md content at install time

2026-03-16 15:45:21 -07:00

test_skill_sources.py

fix(skills): apply PR #491 review feedback (size cap + dedup + conflict mapping)

2026-05-08 13:58:02 -07:00

test_skill_substitution_unification.py

fix(skills): harden task_agent / persona / skill activation from whole-PR review

2026-07-03 19:16:36 -07:00

test_skills_tool.py

fix(skills): address Copilot review — task_agent doc refs + gate fail-closed on missing storage

2026-07-03 19:16:36 -07:00

test_skills.py

refactor(skills): move applied-skill body out of the identity system message

2026-07-03 19:16:36 -07:00

test_sse_chunk_batching.py

feat(sse): coalesce tool_output_chunk emission per call_id

2026-07-20 22:38:32 -07:00

test_sse_cursor_resume.py

fix(compaction): review round 7 — drain crash/order/settle rows, protocol-default fallback, bool event-id guard

2026-07-17 11:28:44 -07:00

test_sse_overflow_js.py

fix(#900): close two vacuity holes the round-2 scenarios left open

2026-07-24 18:15:03 -07:00

test_sse_reconnect_replay.py

fix(sse): make truncated replay recovery lossless and honest

2026-07-20 22:38:32 -07:00

test_sse_recovery_e2e.py

ci: give the e2e_recovery suite its own lane exclusion, drop the live co-mark

2026-07-20 22:38:32 -07:00

test_sse_token_batching.py

feat(sse): coalesce tool_output_chunk emission per call_id

2026-07-20 22:38:32 -07:00

test_state_writer.py

fix(server): apply Copilot + code-quality review feedback

2026-04-25 20:11:47 -07:00

test_storage_attachments.py

feat(preview): rich preview pane + open_preview tool

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

test_storage_notify.py

fix(notify): unbreak PG test backend on the notify dispatcher suite

2026-05-11 01:23:54 -07:00

test_storage_registry.py

Fix/codeql quality findings (#299)

2026-04-04 17:53:20 -07:00

test_storage_skill_spec_uplift.py

chore(skills): drop Anthropic attribution from SKILL.md spec references

2026-05-24 14:56:43 -07:00

test_storage_skills_filtered.py

feat(coordinator): phase 7 — governance + skill metadata + cross-cutt… (#383)

2026-04-18 10:20:19 -07:00

test_storage_source.py

refactor(storage): drop the dead _reminders column instead of carrying it

2026-06-04 11:03:13 -07:00

test_storage_sqlite.py

feat(preview): rich preview pane + open_preview tool

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

test_structured_memory_storage.py

fix(memory): atomic single-statement upsert for memory save/update (#735)

2026-06-28 20:24:20 -07:00

test_structured_memory.py

fix(memory): atomic single-statement upsert for memory save/update (#735)

2026-06-28 20:24:20 -07:00

test_substitute_skill_args.py

fix(skills): harden task_agent / persona / skill activation from whole-PR review

2026-07-03 19:16:36 -07:00

test_think_tag_split.py

refactor(session): extract think-tag splitting into ThinkTagSplitter

2026-08-04 04:53:17 -07:00

test_thumbnails.py

fix(attachments): normalize EXIF orientation so thumbnails and models see upright images

2026-06-16 03:41:51 -07:00

test_tls_admin.py

fix: remove non-auth support from bootstrap wizard (#274)

2026-04-01 19:38:24 -07:00

test_tls_client.py

fix(tls): stub backoff via a _sleep seam, not the global asyncio.sleep

2026-06-16 17:10:12 -07:00

test_tls_manager.py

fix: remove non-auth support from bootstrap wizard (#274)

2026-04-01 19:38:24 -07:00

test_tls_san_renewal.py

fix(tls): repair cluster mTLS — cert identity, renewal scoping, hot-reload

2026-05-30 16:27:15 -07:00

test_tls_storage.py

feat: TLS storage backend + config for lacme integration (#178)

2026-03-25 16:07:05 -07:00

test_tool_advisory.py

feat(operator-context): deliver structured per-kind meta to the UI

2026-06-04 11:03:13 -07:00

test_tool_policies_enforcement.py

feat: replace Redis MQ with direct HTTP transport (Phase 1)

2026-03-30 20:30:05 -07:00

test_tool_policy.py

feat: MCP resource and prompt discovery with read_resource tool (#44)

2026-03-12 14:49:58 -07:00

test_tool_search.py

fix(tool-search): discovery-failure records are per-user, rerank counts honest

2026-08-02 01:48:30 -07:00

test_tool_truncation.py

fix(session): survive mid-stream transport deaths in interactive turns (#937)

2026-08-04 04:53:17 -07:00

test_tools_schema.py

feat(coordinator): MCP tool surface for coordinator sessions (#725)

2026-07-20 07:25:56 -07:00

test_trajectory_model.py

feat(core): add the canonical Turn trajectory model

2026-06-04 11:03:13 -07:00

test_trajectory.py

feat(operator-context): deliver structured per-kind meta to the UI

2026-06-04 11:03:13 -07:00

test_user_storage.py

feat: add PostgreSQL CI integration tests (#156)

2026-03-23 11:11:40 -07:00

test_validate_regex_pattern.py

fix(deadline): prefer a ready result over a same-window deadline/cancel

2026-06-16 17:10:12 -07:00

test_watch_dispatch.py

feat(bash): opt-in background shells with delta output reader and kill tool (#817)

2026-07-10 15:42:33 -07:00

test_watch_integration.py

feat(models): per-alias backend auth via Entra OBO and app identity (#898)

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

test_watch_storage.py

fix(watch): deliver terminal fires instead of dropping them silently

2026-05-14 15:13:00 -07:00

test_watch.py

fix(nudge): log refused wakes; correct the already-dispatched hold-clear comment

2026-07-07 16:42:36 -07:00

test_web_helpers.py

chore: download vendored JS files

2026-07-20 07:21:23 -07:00

test_web_search.py

fix(rerank): harden web_search rerank fallback and correct stale docs

2026-06-01 11:01:30 -07:00

test_webui_auto_approve_visibility.py

fix(test): eliminate lost-wakeup race in approval-prompt tests

2026-07-03 19:16:16 -07:00

test_webui_content.py

test(sse): normalize session_ui_base imports to a single style

2026-07-07 23:32:20 -07:00

test_wire_payload_golden.py

fix(providers): align GPT-5.6 with the GA API surface

2026-07-10 15:59:40 -07:00

test_workstream_endpoints.py

fix(#894): cold flights key on None; typed generation access; abort-Set producer pins

2026-07-24 15:04:31 -07:00

test_workstream_kind.py

feat(coordinator): MCP tool surface for coordinator sessions (#725)

2026-07-20 07:25:56 -07:00