kshitijk4poor 4887aba342 fix(aux): stop republishing the legacy mirrors from a key rotation ...

The rotation tried to decide whether the current runtime was the one
the legacy mirrors were published from, first by key equality and then
by route-tuple equality. Neither identifies the publisher: a scoped
copy with an identical route still matched, and another session could
publish between the comparison and the write (both reproduced by the
second-model verifier). Nothing needs the republish: no production
code reads the legacy globals directly, and _compat_runtime_main()
ignores them while they equal the compat snapshot, which an unchanged
mirror always does. Rotate the published runtime dict in place only,
as before this PR's mirror handling, and pin that an unchanged mirror
never becomes a runtime input.

2026-10-07 13:20:42 +05:30

..

lsp

fix(lsp): no automatic trust for scheduled work; lock-free trust lookups

2026-09-29 03:28:46 +05:30

monitoring

fix(monitoring): report cron store health across every served profile

2026-10-07 10:20:18 +05:30

pet

fix(gateway): deliver and bound-sync generated media in cache/generated

2026-10-02 01:18:03 +05:30

proxy_sources

…

secret_sources

fix(secret-sources): revoke a removed source's value from os.environ

2026-09-29 22:57:48 -04:00

transports

refactor(providers): one copy each of TokenHub's effort rule and the generic nested reasoning fallback

2026-10-07 11:27:35 +05:30

vault_backends

refactor(approval): one "no user can answer" helper for every CLI-callback prompt

2026-10-05 20:08:11 +05:30

verify

fix(processes): signal a process group only when the child leads it

2026-10-05 16:42:57 -05:00

__init__.py

…

account_usage_cache.py

feat(usage): report credential-pool usage per account

2026-10-05 14:02:38 -05:00

account_usage.py

feat(usage): report credential-pool usage per account

2026-10-05 14:02:38 -05:00

acp_openai_bridge.py

…

activity_tracking.py

…

agent_init_fallback.py

fix(copilot): recompute api_mode on fallback and stop stale-mode veto (#46527)

2026-10-06 23:54:51 -05:00

agent_init.py

fix(copilot): recompute api_mode on fallback and stop stale-mode veto (#46527)

2026-10-06 23:54:51 -05:00

agent_runtime_helpers_dead_connections.py

refactor(agent): move dead-connection sweep into a runtime-helpers sibling

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

agent_runtime_helpers.py

refactor(agent): move dead-connection sweep into a runtime-helpers sibling

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

AGENTS.md

feat(i18n): pluggable, layered language packs across core, Desktop and TUI (#126296)

2026-09-28 14:16:18 -07:00

anthropic_adapter.py

fix(lint): stdin guard catches calls whose arguments start on the next line (review)

2026-10-07 01:03:03 +05:30

anthropic_credentials.py

fix(auth): find the Claude CLI outside PATH in external-process and setup-token checks

2026-10-04 17:51:24 -07:00

anthropic_endpoints.py

…

anthropic_message_convert.py

fix(anthropic): price the carrier the converter actually replays

2026-10-02 00:18:00 +05:30

anthropic_thinking_policy.py

fix(anthropic): future Claude ids keep prior thinking by default

2026-10-02 00:18:00 +05:30

anthropic_thinking_replay.py

fix(anthropic): future Claude ids keep prior thinking by default

2026-10-02 00:18:00 +05:30

api_error_summary.py

…

api_request_hooks.py

…

async_utils.py

…

aux_accounting.py

feat(fast): /fast ultrafast for GPT-6 Astra, priced at the served tier

2026-09-29 23:54:34 +05:30

auxiliary_async_rebuild.py

…

auxiliary_call_outcome.py

fix(telemetry): record auxiliary model calls with their real outcome

2026-10-03 13:38:29 -07:00

auxiliary_client_registry.py

feat(auth): brokered PKCE, expiry-driven rotation and aux on plugin OAuth rows

2026-10-05 23:33:41 -07:00

auxiliary_client.py

refactor(aux): move rotate_runtime_main_api_key into auxiliary_key_rotation to keep auxiliary_client.py under its size cap

2026-10-07 13:20:42 +05:30

auxiliary_codex_response.py

refactor(auxiliary): move the Codex aux response normalizer into a sibling

2026-10-05 19:35:01 +05:30

auxiliary_fallback_recovery.py

…

auxiliary_health.py

…

auxiliary_hooks.py

…

auxiliary_key_rotation.py

fix(aux): stop republishing the legacy mirrors from a key rotation

2026-10-07 13:20:42 +05:30

auxiliary_local_runtime.py

fix(fallback): a bare llama.cpp alias resolves only to the live managed server

2026-10-02 16:44:56 -07:00

auxiliary_plugin_oauth.py

feat(auth): brokered PKCE, expiry-driven rotation and aux on plugin OAuth rows

2026-10-05 23:33:41 -07:00

auxiliary_reasoning_floor.py

…

auxiliary_stream_watchdog.py

fix(aux): the first-token window honours providers.<id>.stale_timeout_seconds

2026-10-05 16:00:18 -07:00

auxiliary_structured_output.py

…

auxiliary_task_config.py

refactor: move aux-slot growth into topical siblings (code-health caps)

2026-10-05 17:24:40 -07:00

auxiliary_unavailable.py

fix(fallback): a bare llama.cpp alias resolves only to the live managed server

2026-10-02 16:44:56 -07:00

auxiliary_wire.py

…

azure_identity_adapter.py

…

backend_identity.py

…

background_review.py

fix(agent): cancel background review when session ends or is stopped

2026-09-30 22:59:48 -05:00

battery.py

…

bedrock_adapter.py

fix(bedrock): route GPT-6 family and prefixed OpenAI inference profiles, size their 1M window (#133573)

2026-10-06 08:13:51 -04:00

billing_links.py

…

billing_usage.py

…

billing_view.py

…

bounded_response.py

…

browser_provider.py

…

browser_registry.py

…

chat_completion_helpers_relay.py

…

chat_completion_helpers.py

refactor(providers): LM Studio and Tencent TokenHub get provider profiles; base_url-only agents get snapshot cells

2026-10-07 11:27:35 +05:30

chat_completion_nonstream.py

…

chat_completion_stream_monitor.py

…

chat_completion_wait_notice.py

…

client_lifecycle.py

refactor(compression): import the codex cooldown lookup at module level

2026-10-07 13:20:42 +05:30

codex_headers.py

…

codex_responses_adapter.py

fix(auxiliary): let the shared normalizer judge tool-call-like aux text

2026-10-05 19:35:01 +05:30

codex_runtime_history_seed.py

…

codex_runtime.py

fix(codex): hermes chat -q codex exec/apply_patch approvals fail closed at once

2026-10-05 20:08:11 +05:30

coding_context.py

fix(git): neutralize repository-named filters in the session snapshot and subagent worktrees

2026-10-02 01:24:41 +05:30

command_token_source.py

fix(lint): stdin guard catches calls whose arguments start on the next line (review)

2026-10-07 01:03:03 +05:30

compaction_display.py

…

compression_facade.py

refactor(compression): forward snapshot_is_current directly in facade

2026-10-03 16:54:06 +05:30

compression_marker.py

refactor(agent): share the marker count fragment and retarget the old regex comment

2026-10-02 00:50:20 +05:30

context_breakdown.py

feat(i18n): pluggable, layered language packs across core, Desktop and TUI (#126296)

2026-09-28 14:16:18 -07:00

context_compressor_prellm.py

refactor(compaction): own the bench probe deadline load in PreLlmSkipMixin

2026-10-06 12:03:52 +05:30

context_compressor_summary.py

fix(compression): a keyless pinned route clears the main credential

2026-10-05 15:37:17 -07:00

context_compressor.py

refactor(compaction): name the pre-LLM summary skip in the fallback log

2026-10-06 12:03:52 +05:30

context_engine.py

…

context_file_sources.py

…

context_pin.py

…

context_references.py

refactor(context-refs): read HOME_CREDENTIAL_DIRS directly for the @file block

2026-10-04 13:53:34 +05:30

conversation_compression_archive.py

fix(compression): a row the repair drops is named, not a reason to fall back to the watermark

2026-10-03 18:17:35 +05:30

conversation_compression_codex.py

refactor(compression): import the codex cooldown lookup at module level

2026-10-07 13:20:42 +05:30

conversation_compression_manual.py

fix(compression): publish the /compress here N tail with a rotated child

2026-10-03 18:22:19 +05:30

conversation_compression_reply_anchor.py

…

conversation_compression_telemetry.py

fix(telemetry): a host-timeout compression cancel counts as failed, not skipped

2026-10-05 16:18:47 -07:00

conversation_compression.py

refactor(compression): import the codex cooldown lookup at module level

2026-10-07 13:20:42 +05:30

conversation_loop.py

fix(agent): count restarts since the last response toward the redirect cap

2026-10-02 00:50:20 +05:30

copilot_acp_client.py

…

credential_persistence.py

…

credential_pool_admin.py

…

credential_pool_model_cooldowns.py

feat(desktop): a rate-limited provider says why and until when in the model pickers

2026-10-05 12:10:01 -05:00

credential_pool_plugin.py

feat(auth): brokered PKCE, expiry-driven rotation and aux on plugin OAuth rows

2026-10-05 23:33:41 -07:00

credential_pool.py

feat(auth): brokered PKCE, expiry-driven rotation and aux on plugin OAuth rows

2026-10-05 23:33:41 -07:00

credential_sources.py

…

credits_tracker.py

…

curator_backup.py

…

curator.py

fix(telemetry): loop metrics count only the user's real backend work and curator passes

2026-09-28 12:43:03 -07:00

deadline.py

fix(lint): stdin guard catches calls whose arguments start on the next line (review)

2026-10-07 01:03:03 +05:30

delegation_context.py

fix(kanban): fence the explicit-board override to unfenced callers

2026-10-02 22:21:27 -05:00

display.py

Clarify: one question shape and one result shape on every surface (skip, cancel, timeout and undelivered are told apart) (#127760)

2026-09-29 18:33:44 +05:30

empty_response_guard.py

…

error_classifier.py

fix(classifier): classify a retired free-route 404 as model_not_found

2026-10-02 22:59:00 -05:00

error_surface.py

fix(desktop): open the rejected provider's key from Update API key

2026-09-29 19:31:49 -05:00

errors.py

…

estop.py

…

fallback_cooldown.py

…

fast_mode.py

refactor(tui_gateway): keep the session.create tier parse out of the oversized handler module

2026-10-04 02:12:57 +05:30

file_safety.py

fix(profiles): keep .docker/.azure/.config/gh/.config/gcloud out of profile exports

2026-10-04 13:53:34 +05:30

gemini_native_adapter.py

feat(auth): brokered PKCE, expiry-driven rotation and aux on plugin OAuth rows

2026-10-05 23:33:41 -07:00

gemini_schema.py

…

history_commentary.py

…

i18n_languages.py

feat(i18n): pluggable, layered language packs across core, Desktop and TUI (#126296)

2026-09-28 14:16:18 -07:00

i18n_layers.py

feat(i18n): pluggable, layered language packs across core, Desktop and TUI (#126296)

2026-09-28 14:16:18 -07:00

i18n.py

feat(i18n): pluggable, layered language packs across core, Desktop and TUI (#126296)

2026-09-28 14:16:18 -07:00

image_eviction_policy.py

…

image_gen_provider.py

fix(gateway): deliver and bound-sync generated media in cache/generated

2026-10-02 01:18:03 +05:30

image_gen_registry.py

…

image_routing.py

docs(image-routing): stop claiming the reverted provider-wide supports_vision probe

2026-10-04 17:54:05 -07:00

image_token_cost.py

…

inline_tool_executors.py

Inline agent tools get the shared argument repair (clarify, todo, memory, session_search) (#127781)

2026-09-29 22:07:30 +05:30

insights.py

…

interrupt_compat.py

…

interrupt_control.py

…

interrupt_scope.py

…

iteration_budget.py

…

jiter_preload.py

…

kanban_stop.py

fix(kanban): gate kanban_schedule like the other lifecycle tools

2026-10-01 09:26:00 -04:00

kanban_turn_recovery.py

fix(kanban): retry a failed worker turn in place; carry the recovered response

2026-09-30 22:30:35 -04:00

lazy_forward.py

…

learn_prompt.py

…

learning_graph_render.py

…

learning_graph.py

fix(learning-graph): don't leak 'external' source to later skills in a root

2026-09-30 23:10:24 -05:00

learning_mutations.py

…

legacy_cli.py

…

lmstudio_reasoning.py

refactor(providers): one copy each of TokenHub's effort rule and the generic nested reasoning fallback

2026-10-07 11:27:35 +05:30

manual_compression_feedback.py

…

markdown_tables.py

…

memory_manager.py

…

memory_provider.py

…

message_content.py

…

message_metadata.py

fix(compression): a row the repair drops is named, not a reason to fall back to the watermark

2026-10-03 18:17:35 +05:30

message_sanitization.py

fix(agent): write normalized call_id through transport ToolCall provider_data

2026-10-02 12:29:16 -04:00

micro_compaction.py

fix(compression): name the main model when aux summary falls back

2026-10-05 15:37:17 -07:00

moa_alternation.py

…

moa_loop.py

fix(agent): stamp turn-liveness activity from the MoA reference fan-out

2026-10-02 20:28:55 -05:00

moa_trace.py

…

model_metadata_http.py

…

model_metadata.py

fix(agent): skip /v1/models/{model} probe for unrecognised server types

2026-10-02 20:29:34 -05:00

models_dev.py

…

moonshot_schema.py

…

native_compaction.py

refactor(compaction): drop a dead budget guard and read the user content once

2026-10-05 20:01:03 +05:30

notification_presentation.py

…

nous_rate_guard.py

…

nous_wire.py

…

onboarding.py

…

oneshot_footprint.py

…

oneshot.py

…

opencode_affinity.py

…

outbound_webhooks.py

…

periodic_scheduler.py

…

plan_prompt.py

…

plugin_llm.py

…

plugin_stream_hooks.py

…

portal_tags.py

…

process_bootstrap.py

…

prompt_builder.py

feat(agent): truncated context files name the sections that were dropped

2026-10-04 23:05:42 -07:00

prompt_cache_boundary.py

…

prompt_cache_scope.py

Free tier: a rate limit ends a watched turn at once with the reset time (#134049)

2026-10-06 12:31:21 -07:00

prompt_caching.py

refactor(cache): one scaffold-split builder and one prior-decoration strip pass

2026-10-06 14:01:25 +05:30

provider_base.py

…

provider_media.py

fix(gateway): deliver and bound-sync generated media in cache/generated

2026-10-02 01:18:03 +05:30

provider_projection.py

…

provider_registry.py

…

proxy_bypass.py

…

rate_limit_credits.py

…

rate_limit_tracker.py

…

reactions.py

…

reasoning_effort.py

refactor(providers): one copy each of TokenHub's effort rule and the generic nested reasoning fallback

2026-10-07 11:27:35 +05:30

reasoning_params.py

…

reasoning_promotion.py

refactor(agent): drop redundant str() wrapping around custom_provider_slug args

2026-10-06 13:01:39 +05:30

reasoning_summaries.py

…

reasoning_timeouts.py

…

redact.py

refactor(debug): reuse agent.redact lists in support redaction

2026-10-03 21:15:17 +05:30

relay_cwd.py

…

relay_llm.py

fix(relay): native Anthropic stream spans keep their input and cache token counts

2026-10-07 00:18:10 +05:30

relay_runtime.py

fix(telemetry): record auxiliary model calls with their real outcome

2026-10-03 13:38:29 -07:00

relay_tools.py

…

repetition_guard.py

refactor(agent): drop the stream watch's redundant pre-trim

2026-10-02 00:53:19 +05:30

replay_cleanup.py

…

retry_utils.py

Free tier: a rate limit ends a watched turn at once with the reset time (#134049)

2026-10-06 12:31:21 -07:00

review_engine.py

…

review_idle_queue.py

…

runtime_cwd.py

…

runtime_self_protection.py

…

sdk_transform_bypass.py

…

search_policy.py

fix(agent): a home working dir is a fallback, not a project

2026-09-29 14:53:00 -05:00

secret_scope.py

fix(secret-scope): compose the managed .env into every profile secret scope

2026-09-29 22:57:48 -04:00

served_model.py

…

session_activity.py

fix(cron): preserve suspend credit across idle sampling

2026-10-02 13:50:23 -04:00

session_persistence.py

perf(sessions): pair tool-call uids in O(K) on flush and only when needed on restore

2026-09-29 04:15:21 +05:30

session_source.py

Free tier: a rate limit ends a watched turn at once with the reset time (#134049)

2026-10-06 12:31:21 -07:00

shell_hooks.py

fix(hooks): block failed hooks without directives

2026-10-02 20:18:26 -05:00

side_question.py

…

skill_bundles.py

…

skill_commands.py

refactor(skills): tag the plugin-skill projection instead of invalidating from plugins.py

2026-10-05 00:23:05 -07:00

skill_preprocessing.py

fix(skills): trust-scope inline-shell auto-exec to non-community hub skills

2026-10-06 11:20:43 -04:00

skill_utils.py

refactor: remove Skill Sync (hermes sync) and the org skill mirror from core

2026-10-03 23:05:48 -07:00

ssl_verify.py

docs(ssl): note SSL_CERT_FILE sits on top of the platform store

2026-10-06 12:25:01 +05:30

status_output.py

fix(memory): show every memory-provider migration outcome, ask once per update

2026-10-02 04:09:57 -07:00

stream_delivery.py

…

stream_diag.py

…

stream_single_writer.py

…

subagent_lifecycle.py

…

subdirectory_hints.py

fix(agent): a home working dir is a fallback, not a project

2026-09-29 14:53:00 -05:00

subscription_view.py

…

surface_switch.py

…

system_prompt.py

fix(agent): gate execution-guidance tool lines on the session's toolset

2026-10-01 11:45:41 -07:00

terminal_approval_batch.py

refactor(security): remove the bundled tirith scanner from core

2026-10-06 14:22:13 -07:00

terminal_env_provider.py

…

terminal_env_registry.py

…

think_scrubber.py

…

thinking_timeout_guidance.py

…

thread_scoped_output.py

…

title_generator.py

…

tool_call_batches.py

fix(agent): execute local tool batches through per-tool pipeline

2026-10-01 23:50:31 -07:00

tool_dispatch_helpers.py

fix(tools): an image-bearing MCP result keeps its incomplete-data notice (review)

2026-10-06 22:54:13 -07:00

tool_executor.py

fix(tools): image-bearing tool results stay inside the turn budget; spills and refusals keep their content (review)

2026-10-06 22:54:13 -07:00

tool_guardrails.py

feat(telemetry): harness-accuracy shared metrics for the agent loop

2026-09-28 12:43:03 -07:00

tool_result_classification.py

…

trace_upload.py

…

trajectory.py

…

transcript_repair.py

fix(state): keep stored token_count when the agent flush row carries token_count=None

2026-10-02 01:04:12 +05:30

transcription_provider.py

feat(stt): live transcription while you speak on CLI, TUI and Desktop (stt.streaming)

2026-10-06 12:50:31 -07:00

transcription_registry.py

…

tts_provider.py

fix(tts): validate a plugin's stream rate once; keep nous off plugin routing

2026-10-06 00:52:41 -07:00

tts_registry.py

…

turn_api_call.py

fix(desktop): mark replies the user stopped as "Response stopped"

2026-10-01 03:27:39 -05:00

turn_api_error.py

Free tier: a rate limit ends a watched turn at once with the reset time (#134049)

2026-10-06 12:31:21 -07:00

turn_api_request.py

feat(telemetry): v5 efficiency metrics — turn cost, wasted tokens, tool output truncation, tool overhead, prompt-cache breaks

2026-09-28 12:43:03 -07:00

turn_author.py

…

turn_context_compaction.py

…

turn_context.py

fix(agent): name a final that reuses an already-delivered response

2026-10-04 18:04:17 -07:00

turn_empty_response.py

fix(agent): gate reasoning-only answer promotion by route

2026-10-06 13:01:39 +05:30

turn_explainers.py

feat(i18n): pluggable, layered language packs across core, Desktop and TUI (#126296)

2026-09-28 14:16:18 -07:00

turn_facade_lease.py

refactor(state): hoist pidns imports and trim duplicated call-site prose

2026-10-03 17:44:58 +05:30

turn_facade.py

fix: use original gateway text for automatic session titles

2026-09-29 15:57:52 -05:00

turn_failure_copy.py

Free tier: a rate limit ends a watched turn at once with the reset time (#134049)

2026-10-06 12:31:21 -07:00

turn_final_response.py

refactor(agent): derive reasoning-promotion trust live in agent/reasoning_promotion

2026-10-06 13:01:39 +05:30

turn_finalizer.py

fix(agent): name a final that reuses an already-delivered response

2026-10-04 18:04:17 -07:00

turn_iteration_prep.py

Free tier: a rate limit ends a watched turn at once with the reset time (#134049)

2026-10-06 12:31:21 -07:00

turn_liveness.py

fix(agent): stamp the activity clock at turn end and dump thread stacks on watchdog fire

2026-10-03 16:18:04 -05:00

turn_loop_errors.py

…

turn_overflow.py

…

turn_preflight_gate.py

…

turn_preflight.py

…

turn_recovery_autorecover.py

Free tier: a rate limit ends a watched turn at once with the reset time (#134049)

2026-10-06 12:31:21 -07:00

turn_recovery.py

fix(agent): a credit-limited 402 retries once with the affordable output cap instead of failing as billing

2026-10-07 06:41:07 +05:30

turn_request_assembly.py

fix(anthropic): simplify recovery and make accounting explicit

2026-10-02 00:18:00 +05:30

turn_response_check.py

feat(telemetry): harness-accuracy shared metrics for the agent loop

2026-09-28 12:43:03 -07:00

turn_response_intake.py

feat(telemetry): harness-accuracy shared metrics for the agent loop

2026-09-28 12:43:03 -07:00

turn_retry_state.py

fix(agent): a credit-limited 402 retries once with the affordable output cap instead of failing as billing

2026-10-07 06:41:07 +05:30

turn_stop_gates.py

…

turn_summary.py

…

turn_tool_round.py

fix(agent): normalize the final staged tool batch by effective id

2026-10-02 12:29:16 -04:00

turn_tool_validation.py

fix(agent): normalize provider-minted parallel tool ids

2026-10-02 12:29:16 -04:00

turn_truncation.py

fix(agent): cut a runaway repetition loop while it streams, callback or not

2026-10-02 00:53:19 +05:30

turn_usage.py

Free tier: a rate limit ends a watched turn at once with the reset time (#134049)

2026-10-06 12:31:21 -07:00

usage_anchor.py

…

usage_pricing.py

refactor(pricing): generate the gpt-6.1-sol -pro/-900k aliases from the tier loop

2026-10-01 16:40:58 +05:30

vault_login_classifier.py

…

vault_store.py

…

verification_evidence.py

fix(verify): share only a full passing hermes verify across sessions

2026-10-05 16:43:19 -05:00

verification_stop.py

…

verify_hooks.py

…

vertex_adapter.py

…

video_gen_provider.py

fix(gateway): deliver and bound-sync generated media in cache/generated

2026-10-02 01:18:03 +05:30

video_gen_registry.py

…

vision_message_prep.py

…

web_search_provider.py

…

web_search_registry.py

fix(web): pick own key or Nous Tool Gateway per web capability in Desktop

2026-10-06 02:53:56 -07:00