Voon Foo 2f24d4457e fix(ai,catalog): widen Bedrock stream-stall watchdog via model compat ...

The lazy provider wrapper ignored model.compat.streamIdleTimeoutMs, so
Bedrock reasoning models sat on the generic 300s idle watchdog despite
ConverseStream sending no ping keepalives; long quiet thinking runs died
with "Provider stream stalled while waiting for the next event" during
plan writing and todo execution (issue #4758's Bedrock variant, worst on
Fable 5 where the display default flipped to omitted).

- catalog: BedrockCompat gains streamIdleTimeoutMs; reasoning models get
  a 600s floor, adaptive-thinking Claude (Opus 4.7+, Sonnet/Opus 5,
  Fable/Mythos 5) 900s to match direct Anthropic's ping-extended
  tolerance; explicit compat overrides still win (0 disables).
- ai: forwardStream resolves options -> env -> model.compat -> default,
  and lazy terminal errors carry the structural errorId classification
  so session auto-retry classifies stalls without text matching.

2026-08-07 14:18:46 +08:00

..

data

feat(ai): add @oh-my-pi/pi-ai package with tool name prefixing

2026-01-09 12:10:03 +01:00

fixtures

refactor(coding-agent): deferred startup parsing and schema compilation

2026-08-03 16:31:27 +02:00

helpers

feat: enabled codex responses lite for gpt-5.6 models and remote compaction

2026-07-10 09:46:27 +02:00

providers

chore: reorg tests

2026-07-27 16:43:53 +02:00

registry/oauth

chore: reorg tests

2026-07-27 16:43:53 +02:00

abort-source-tracker.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

abort.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

alibaba-endpoint-selection.test.ts

fix(ai): validated custom alibaba endpoints by model list

2026-07-21 20:53:55 +00:00

alibaba-token-plan-usage.test.ts

fix(ai): repair Alibaba Token Plan usage reporting

2026-07-24 14:00:50 +00:00

alibaba-token-plan.test.ts

Merge PR #6683: fix(providers): add China (Beijing) region for alibaba-token-plan (@roboomp)

2026-07-26 15:41:26 +02:00

anthropic-abandoned-tooluse-replay.test.ts

fix(ai): scope demoted-thinking separator to the demoted text block itself

2026-07-03 23:35:13 +00:00

anthropic-alignment.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

anthropic-client.test.ts

test(ai): prove Anthropic stream rejection

2026-07-30 04:09:15 +09:00

anthropic-copilot-checkpoint-thinking-signature.test.ts

fix(ai): restore the drop guard for undefined thinking signatures

2026-07-03 21:48:21 +09:00

anthropic-empty-error-tool-result.test.ts

fix: address review comments — named AnthropicToolResultContent type, biome formatting

2026-06-10 12:54:11 +02:00

anthropic-error-tool-result-image.test.ts

fix(ai): hoisted images out of error Anthropic tool results

2026-06-12 11:07:35 +02:00

anthropic-fable-request-shaping.test.ts

fix(providers): mapped minimax direct efforts

2026-06-18 05:28:31 +00:00

anthropic-fast-mode.test.ts

fix(rpc): report Anthropic fast fallback honestly

2026-07-29 20:58:01 +01:00

anthropic-many-image-resize.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

anthropic-mid-conversation-system.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

anthropic-oauth.test.ts

fix(ai): surfaced inactive Claude scoped weekly limits in omp usage

2026-07-26 16:12:51 +02:00

anthropic-ping-keepalive.test.ts

fix(ai): bounded anthropic ping keepalive extension of idle watchdog

2026-07-09 18:27:20 +02:00

anthropic-prefill.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

anthropic-prior-turn-thinking.test.ts

fix(ai): strip foreign thinking signatures on the latest assistant turn

2026-07-23 20:41:43 +09:00

anthropic-retry.test.ts

fix(ai): retry GitHub Copilot fleet-skew model 400s

2026-08-04 21:45:44 +01:00

anthropic-server-side-fallback.test.ts

fix(ai): kept assistant tool_use blocks trailing on Anthropic replay

2026-07-07 12:06:02 -03:00

anthropic-signature-auto-mark.test.ts

feat(ai): auto-marked signing anthropic-messages proxies from 400

2026-07-02 11:54:47 +00:00

anthropic-signature-hint.test.ts

style: bun run fix

2026-07-02 12:09:16 +00:00

anthropic-stream-envelope.test.ts

fix(ai): addressed anthropic stream truncation errors for tool call recovery

2026-08-02 06:14:01 +02:00

anthropic-stream-timeout.test.ts

fix(ai): honor model.compat.streamIdleTimeoutMs in the Anthropic idle watchdog

2026-08-01 15:15:31 +08:00

anthropic-thinking-immutability.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

anthropic-thinking-only-length-truncated.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

anthropic-tool-schema.test.ts

Fix Anthropic root tool schema combinators

2026-06-17 22:06:51 +08:00

anthropic-unsigned-thinking-replay.test.ts

feat(ai): standardized thinking block demotion across model providers

2026-06-27 08:19:11 +02:00

anthropic-zenmux-checkpoint-thinking-signature.test.ts

fix(ai): restore the drop guard for undefined thinking signatures

2026-07-03 21:48:21 +09:00

api-key-validation.test.ts

fix(ai): preserve API-key validation HTTP errors

2026-07-29 23:09:07 +02:00

api-registry.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

apply-patch-freeform.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

auth-broker-config-discovery.test.ts

fix(ai): reject padded account pool identity keys

2026-07-23 21:37:51 +00:00

auth-broker-disabled-credentials.test.ts

feat: implemented credential lifecycle tracking and management APIs

2026-07-25 18:02:43 +02:00

auth-broker-nested-config.test.ts

fix(auth-broker): preserve config.yaml discovery

2026-07-14 18:33:57 +02:00

auth-broker-oauth-extra-fields.test.ts

fix(pr-3347): use worktree-local cleanup helper in ai tests

2026-06-27 02:04:51 +02:00

auth-broker-refresher.test.ts

fix(pr-3347): use worktree-local cleanup helper in ai tests

2026-06-27 02:04:51 +02:00

auth-broker-remote-store.test.ts

feat: implemented client usage tracking and reporting in the auth broker

2026-07-24 15:20:46 +02:00

auth-broker-snapshot-cache.test.ts

fix(ai): preserve Codex broker block compatibility

2026-07-29 23:08:10 +02:00

auth-broker-wire-schema-contract.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

auth-broker-wire.test.ts

fix(ai): filter unrelated auth database changes

2026-07-29 23:08:11 +02:00

auth-gateway-anthropic-caching.test.ts

test: replaced real timers and sleeps with deterministic test hooks

2026-05-17 04:02:09 +02:00

auth-gateway-anthropic-messages.test.ts

fix(ai): validated preserved web-search history blocks

2026-07-24 11:07:13 +00:00

auth-gateway-anthropic-to-codex-caching.test.ts

test: replaced real timers and sleeps with deterministic test hooks

2026-05-17 04:02:09 +02:00

auth-gateway-cache-key.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

auth-gateway-classify-error.test.ts

refactor: centralized AI error handling into a dedicated module

2026-06-27 10:44:13 +02:00

auth-gateway-cross-protocol-caching.test.ts

test: replaced real timers and sleeps with deterministic test hooks

2026-05-17 04:02:09 +02:00

auth-gateway-model-list.test.ts

fix(ai): qualified auth-gateway model ids

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

auth-gateway-openai-chat.test.ts

feat(openai): add explicit prompt cache policy

2026-07-23 15:18:37 -04:00

auth-gateway-openai-prompt-cache.test.ts

feat(openai): add explicit prompt cache policy

2026-07-23 15:18:37 -04:00

auth-gateway-openai-responses-caching.test.ts

test(tests): updated test setup to use shared E2E helpers and reset settings

2026-05-26 20:51:03 +02:00

auth-gateway-openai-responses.test.ts

feat: add native Codex computer use

2026-07-24 01:40:04 +00:00

auth-gateway-pi-native.test.ts

Merge PR #6413: feat(cli): benchmark prompt cache reuse (@riverpilot)

2026-07-24 02:25:24 +02:00

auth-gateway-response-headers.test.ts

feat(auth-gateway): added diagnostic headers to auth-gateway inference responses

2026-07-13 18:48:20 +02:00

auth-gateway-thinking-loop.test.ts

fix: honor gateway quota usage limits

2026-07-12 03:08:29 -04:00

auth-retry.test.ts

fix(ai): keep concurrency caps out of auth rotation

2026-08-05 02:32:36 +02:00

auth-storage-account-identity.test.ts

fix(pr-3347): use worktree-local cleanup helper in ai tests

2026-06-27 02:04:51 +02:00

auth-storage-antigravity-selection.test.ts

fix(pr-3347): use worktree-local cleanup helper in ai tests

2026-06-27 02:04:51 +02:00

auth-storage-api-key-login.test.ts

feat: add native Alibaba Token Plan provider

2026-07-23 23:04:00 +00:00

auth-storage-block-persistence.test.ts

fix(ai): make Codex block migration atomic

2026-07-29 23:08:11 +02:00

auth-storage-broker-no-sentinel.test.ts

fix(auth): rotate through quota-limited accounts

2026-07-12 00:23:04 -04:00

auth-storage-check-credentials.test.ts

fix(auth): skipped local usage probes in health checks

2026-06-18 07:38:37 +00:00

auth-storage-claude-fable-fallback.test.ts

fix(coding-agent): scope advisor cost to active session

2026-07-29 07:18:48 +00:00

auth-storage-close-releases-handles.test.ts

test(ai): assert close() releases the SQLite connection

2026-07-28 10:58:58 +02:00

auth-storage-codex-selection.test.ts

Merge PR #7618: fix(ai): respect explicit Codex usage allowance (@roboomp)

2026-08-05 02:22:00 +02:00

auth-storage-codex-workspace-identity.test.ts

test(ai): pin codex workspace-scoped identity invariants

2026-07-20 01:15:31 -04:00

auth-storage-config-override.test.ts

fix(pr-3347): use worktree-local cleanup helper in ai tests

2026-06-27 02:04:51 +02:00

auth-storage-credential-disabled-event.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

auth-storage-credential-origin.test.ts

fix(ai): preferred explicit env api key over stored static key

2026-06-30 06:34:07 +02:00

auth-storage-email-dedupe.test.ts

fix(ai): preserve legacy Codex SQLite compatibility

2026-07-29 23:08:11 +02:00

auth-storage-force-refresh-rotate.test.ts

fix: honor gateway quota usage limits

2026-07-12 03:08:29 -04:00

auth-storage-manual-code-gate.test.ts

fix(ai): 仅为 paste-code provider 合成默认手动粘贴码提示

2026-06-26 16:13:25 +08:00

auth-storage-model-usage-health.test.ts

fix(ai): rejected broker block writes after corruption

2026-08-01 19:13:42 +00:00

auth-storage-oauth-account-select.test.ts

feat(security): OMP-native security subsystem (planner handoff)

2026-07-29 18:47:51 -07:00

auth-storage-oauth-refresh-race.test.ts

fix(ai): refresh expired oauth credential on observed mismatch

2026-07-31 14:04:35 +00:00

auth-storage-org-scoped-identity.test.ts

fix(ai): claim same-org rows via stored credential bases; stable org-only usage cache identity

2026-07-11 22:49:12 +09:00

auth-storage-refresh-skew.test.ts

fix(pr-3347): use worktree-local cleanup helper in ai tests

2026-06-27 02:04:51 +02:00

auth-storage-sqlite-busy.test.ts

test(ai): replaced busy-handler ordering spy with real lock contention

2026-08-01 21:42:05 +02:00

auth-storage-usage-cache.test.ts

fix(ai): repair Alibaba Token Plan usage reporting

2026-07-24 14:00:50 +00:00

auth-storage-usage-history.test.ts

fix(usage): expired opencode cache after cost writes

2026-06-18 07:52:45 +00:00

auth-storage-xai-oauth-usage.test.ts

fix(ai): bound SuperGrok OAuth usage

2026-07-21 23:21:32 -03:00

auth-storage-zai-api-key-selection.test.ts

refactor(usage): renamed Z.AI feature quota label and tier identifier

2026-07-24 14:06:34 +02:00

aws-credentials.test.ts

fix(ai): address Bedrock Mantle review feedback

2026-08-01 12:29:07 +03:00

aws-eventstream.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

aws-registry.test.ts

fix(ai): address Bedrock Mantle review feedback

2026-08-01 12:29:07 +03:00

aws-sigv4.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

azure-openai-responses-stream.test.ts

fix(ai): honor Azure runtime endpoint for computer tools

2026-07-25 20:47:33 +00:00

bedrock-inference-profile.test.ts

Merge PR #7080: fix(ai): add authenticated Bedrock Mantle routing (@anatoli-tsinovoy)

2026-08-03 14:36:52 +02:00

bedrock-mantle-auth.test.ts

fix(ai): address Bedrock Mantle review feedback

2026-08-01 12:29:07 +03:00

bedrock-prompt-cache.test.ts

test(ai): scope Bedrock auth bypass to stream calls

2026-07-31 19:07:16 +02:00

bedrock-system-prompt.test.ts

test(ai): scope Bedrock auth bypass to stream calls

2026-07-31 19:07:16 +02:00

callback-server-launch-route.test.ts

fix(ai): clarified oauth completion close copy

2026-07-09 17:25:23 +00:00

callback-server-manual-input.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

callback-server-port-fallback.test.ts

test(ai): occupy 127.0.0.1 explicitly in the port-fallback blocker

2026-07-23 22:15:22 +02:00

callback-server-security.test.ts

fix(ai): surface state-authenticated provider denials instead of timing out

2026-07-23 22:15:22 +02:00

claude-ratelimit-headers.test.ts

feat(ai): integrated granular credential usage tracking and rate limiting

2026-07-02 22:38:54 +02:00

claude-usage-headers.test.ts

fix(ai): surfaced inactive Claude scoped weekly limits in omp usage

2026-07-26 16:12:51 +02:00

claude-usage-retry.test.ts

fix(ai): separated anthropic account and organization identity

2026-07-17 07:55:41 +00:00

context-overflow.test.ts

refactor: centralized AI error handling into a dedicated module

2026-06-27 10:44:13 +02:00

copilot-retry.test.ts

fix(ai): scope the Copilot 8-attempt retry budget to model flaps

2026-08-04 21:51:35 +01:00

coreweave-login.test.ts

fix(coreweave): harden project header setup

2026-07-01 10:49:01 -04:00

coreweave-project-header.test.ts

fix(coreweave): drop blank project header overrides

2026-07-01 11:02:39 -04:00

cursor-exec-handlers.test.ts

fix(cursor): surfaced K3 missing-thinking turns

2026-08-03 15:26:12 +00:00

cursor-exec-modern.test.ts

fix(cursor): corrected inline read range metadata

2026-08-04 04:13:35 +00:00

cursor-h2-transport-error.test.ts

fix(cursor): surfaced actionable error when h2 ALPN negotiation fails

2026-07-17 10:27:57 +00:00

cursor-mcp-tool-catalog.test.ts

fix(cursor): honor Pi frame arguments and preserve open-block args

2026-07-30 01:41:55 +02:00

cursor-streaming-args.test.ts

feat(ai): handle Cursor's modern exec wire protocol

2026-07-30 01:41:55 +02:00

cursor-terminal-error.test.ts

fix(cursor): pair interrupted calls and gate advisor bridge approval

2026-07-30 01:42:28 +02:00

cursor-todo-bridge.test.ts

feat(ai): handle Cursor's modern exec wire protocol

2026-07-30 01:41:55 +02:00

cursor-transport-error.test.ts

test(ai,catalog): give spawn-based lazy tests explicit timeouts

2026-07-29 23:09:06 +02:00

cursor-usage.test.ts

test(ai): cover Cursor usage registration

2026-07-14 19:16:09 +02:00

deepseek-reasoning-content.test.ts

test(ai): aligned deepseek flash ladder expectations with #7668

2026-08-06 14:18:50 +02:00

devin-frame-cap.test.ts

style: bun run fix

2026-07-02 08:32:05 +00:00

devin-history.test.ts

test(ai): cover bare-string Devin system prompt

2026-07-29 23:09:13 +02:00

devin-large-request-recovery.test.ts

fix(devin): retain flushed history in overflow sizing

2026-07-20 13:37:36 -04:00

devin-login.test.ts

feat: added devin provider support

2026-06-22 04:45:17 +02:00

devin-streaming-args.test.ts

style: apply formatter and fix devin test enum

2026-07-01 04:45:28 +02:00

devin-usage.test.ts

fix(ai): include Devin cache usage in total tokens

2026-07-18 16:40:15 -07:00

dialect-thinking.test.ts

fix(ai/dialect): prevented premature termination of nested thinking blocks

2026-06-30 21:27:12 +02:00

duplicate-tool-results.test.ts

fix(ai/providers): prevented cross-turn frame and tool result misattribution

2026-06-21 08:30:37 +02:00

empty-completion-retry.test.ts

fix(ai): retried empty ollama completions

2026-07-06 02:31:23 +00:00

error-aierr.test.ts

fix(ai): retried statusless capacity errors

2026-07-24 10:30:29 +00:00

error-id.test.ts

test(ai): preserve auth failure precedence

2026-07-20 19:49:09 -05:00

eval-language-whitespace.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

event-stream.test.ts

refactor: centralized AI error handling into a dedicated module

2026-06-27 10:44:13 +02:00

exa-login.test.ts

feat(ai): add Exa API key login

2026-07-25 19:08:02 -04:00

firepass.live.ts

fix(catalog): moved OpenAI effort maps into thinking metadata

2026-06-11 23:28:07 +00:00

firepass.test.ts

fix(catalog): moved OpenAI effort maps into thinking metadata

2026-06-11 23:28:07 +00:00

fireworks-login.test.ts

fix(ai): validated Fireworks login against control-plane catalog

2026-06-22 00:16:31 +00:00

gemini-gemma-dialect.test.ts

feat: implemented shared serialization utilities and standardized inventories

2026-08-03 05:06:49 +02:00

github-copilot-anthropic-auth.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

github-copilot-anthropic-fleet-skew.test.ts

fix(ai): retry GitHub Copilot fleet-skew model 400s

2026-08-04 21:45:44 +01:00

github-copilot-error.test.ts

fix(ai): retry GitHub Copilot fleet-skew model 400s

2026-08-04 21:45:44 +01:00

github-copilot-headers.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

github-copilot-login.test.ts

fix(auth): routed copilot business endpoint

2026-06-17 12:47:19 +00:00

github-copilot-long-context-wire.test.ts

fix(ai): degrade Copilot Responses image detail "original" to auto

2026-06-17 22:38:30 -07:00

github-copilot-openai-base-url.test.ts

fix(ai): omit service tier for GitHub Copilot

2026-07-11 14:40:07 +08:00

github-copilot-reasoning.test.ts

fix(ai): advertised context-management beta for api-key thinking

2026-06-23 01:35:42 +00:00

gitlab-duo-cache-key.test.ts

feat(cli): benchmark prompt cache reuse

2026-07-23 15:18:37 -04:00

gitlab-duo-workflow-oauth.test.ts

feat(ai): add GitLab Duo Agent provider

2026-06-26 16:13:20 +08:00

gitlab-duo-workflow-provider.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

glm-5.2-reasoning-effort.test.ts

feat: introduced max reasoning effort tier across model and rpc systems

2026-07-10 13:39:42 +02:00

google-antigravity-usage.test.ts

fix(ai): preserved near-reset antigravity weekly windows

2026-06-29 09:36:06 +00:00

google-cached-content.test.ts

fix(gateway): preserve Google cached content through pi-native

2026-07-23 16:45:33 -04:00

google-empty-response-retry.test.ts

fix(ai): fail over after empty Antigravity streams

2026-07-21 10:00:52 -05:00

google-function-calling-matching.test.ts

fix(ai): healed leaked reasoning fences live for every provider

2026-06-30 06:38:25 +02:00

google-gemini-cli-3x-thinking.test.ts

fix(gemini): suppress CCA thinking summaries by default

2026-07-04 13:50:19 +08:00

google-gemini-cli-429.test.ts

fix: resolved auth-gateway handling of 429 usage-limit responses

2026-05-28 02:56:54 +02:00

google-gemini-cli-alignment.test.ts

fix(ai): fail over after empty Antigravity streams

2026-07-21 10:00:52 -05:00

google-gemini-cli-variant-routing.test.ts

feat(catalog): enabled budget-aware Antigravity routing and request envelope tracking

2026-06-17 13:47:40 +02:00

google-oauth-hostname.test.ts

fix(oauth): bound and cancel google provisioning poll

2026-07-23 20:08:11 +00:00

google-oauth-validation-url.test.ts

fix(oauth): bound and cancel google provisioning poll

2026-07-23 20:08:11 +00:00

google-service-tier.test.ts

feat(ai): removed legacy Google interactions routing from AI providers

2026-07-13 18:43:52 +02:00

google-system-prompt.test.ts

feat(ai): removed legacy Google interactions routing from AI providers

2026-07-13 18:43:52 +02:00

google-thinking-signature.test.ts

build(config): refactored build and TypeScript configuration with Bun loaders

2026-01-24 00:03:52 +01:00

google-tool-choice.test.ts

fix(ai): omitted google auto tool config

2026-06-16 13:13:27 +00:00

google-tool-schema.test.ts

fix(ai): inspect snake-case schema children

2026-07-30 01:27:35 +02:00

handoff.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

harmony-leak.test.ts

feat(cross-cutting): added multi-syntax in-band tool-call support for runtime tool conversion

2026-06-15 07:33:24 +02:00

http-inspector.test.ts

fix(ai): align dump status extraction

2026-07-05 13:10:25 +02:00

image-limits.test.ts

test(ai): migrate 24 test files from fs.rm to removeWithRetries

2026-06-23 12:43:23 -07:00

image-tool-result.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

inband-tools.test.ts

feat(ai): implemented automated recovery for malformed tool arguments

2026-07-11 14:00:14 +02:00

issue-814-repro.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

issue-826-repro.test.ts

refactor(catalog): baked thinking metadata into buildModel pipeline

2026-06-10 07:22:11 +02:00

issue-827-repro.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

issue-883-repro.test.ts

feat(ai): consolidated OpenAI-family streaming and add OpenRouter API support

2026-06-17 21:36:48 +02:00

issue-911-repro.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

issue-912-repro.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

issue-931-repro.test.ts

feat(ai): consolidated OpenAI-family streaming and add OpenRouter API support

2026-06-17 21:36:48 +02:00

issue-945-repro.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

issue-955-repro.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

issue-957-repro.test.ts

fix(pr-3347): use worktree-local cleanup helper in ai tests

2026-06-27 02:04:51 +02:00

issue-959-repro.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

issue-967-vision-guard.test.ts

fix(ai): omitted unsupported reasoning sampling params

2026-07-15 19:30:53 +00:00

issue-969-repro.test.ts

refactor(catalog): baked thinking metadata into buildModel pipeline

2026-06-10 07:22:11 +02:00

issue-976-repro.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

issue-1203-repro.test.ts

fix(ai): persisted minimax cumulative reasoning dedup across block transitions

2026-06-14 07:14:10 +00:00

issue-1207-repro.test.ts

test(ai): aligned deepseek flash ladder expectations with #7668

2026-08-06 14:18:50 +02:00

issue-1227-repro.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

issue-1270-repro.test.ts

fix: google-vertex support multi-region endpoints

2026-07-27 19:36:48 +02:00

issue-1373-repro.test.ts

refactor(catalog): baked thinking metadata into buildModel pipeline

2026-06-10 07:22:11 +02:00

issue-1417-repro.test.ts

fix(pr-3347): use worktree-local cleanup helper in ai tests

2026-06-27 02:04:51 +02:00

issue-1701-repro.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

issue-1776-repro.test.ts

feat: implemented symbol-based streaming state for isolated metadata

2026-06-27 12:07:08 +02:00

issue-1838-repro.test.ts

feat(ai): consolidated OpenAI-family streaming and add OpenRouter API support

2026-06-17 21:36:48 +02:00

issue-2080-repro.test.ts

feat(ai): consolidated OpenAI-family streaming and add OpenRouter API support

2026-06-17 21:36:48 +02:00

issue-2123-repro.test.ts

refactor(catalog): baked thinking metadata into buildModel pipeline

2026-06-10 07:22:11 +02:00

issue-2315-repro.test.ts

fix(ai): overlay compat.reasoningEffortMap onto thinking.effortMap

2026-06-11 23:37:49 +00:00

issue-2424-repro.test.ts

fix(ai): rejected https gitlab duo loopback redirects

2026-06-14 07:13:07 +00:00

issue-2883-moonshot-base-url.test.ts

fix(ai): honored Moonshot base URL during login

2026-07-18 11:25:34 +00:00

issue-2996-repro.test.ts

test(ai): Added regression test for reasoning_content with null content delta

2026-06-18 21:25:04 -07:00

issue-3124-repro.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

issue-3434-repro.test.ts

fix(ai): route Bedrock dotted Claude profiles to Anthropic dialect and separate flattened bare demoted-thinking blocks

2026-07-03 20:54:35 +00:00

issue-3528-repro.test.ts

fix(ai): route Bedrock dotted Claude profiles to Anthropic dialect and separate flattened bare demoted-thinking blocks

2026-07-03 20:54:35 +00:00

issue-3555-repro.test.ts

fix(ai): surfaced ollama usage accounts

2026-06-26 09:59:12 +00:00

issue-3593-repro.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

issue-4085-repro.test.ts

test: replaced banned Promise executor with withResolvers in 4085 repro

2026-07-24 02:24:16 +02:00

issue-4593-repro.test.ts

fix(ai): deferred stream idle watchdog while cursor exec tools run locally

2026-07-09 18:30:48 +02:00

issue-4679-repro.test.ts

fix(providers): disabled strict tools on azure anthropic

2026-07-06 06:19:48 +00:00

issue-5983-repro.test.ts

fix(catalog): normalized kimi k3 reasoning efforts

2026-07-18 11:36:38 +00:00

issue-6510-repro.test.ts

fix(providers): aligned zen context-management beta

2026-07-24 12:05:18 +00:00

issue-6913-harmony-marker-escaping.test.ts

feat: use simple form replace when replace is the edit mode

2026-08-03 01:00:51 +02:00

json-schema-typescript.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

kagi-login.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

kilo-login.test.ts

feat: added injectable fetch transport across AI and coding network flows

2026-06-09 04:51:17 +02:00

kimi-usage.test.ts

fix(kimi): surface the 5h usage window reset time

2026-07-18 11:48:20 +08:00

leaked-thinking-stream.test.ts

fix(ai): reset leaked text signatures by source

2026-07-26 13:54:42 +00:00

litellm-login.test.ts

fix(litellm): support LITELLM_BASE_URL

2026-06-16 20:54:55 -04:00

max-effort-wire.test.ts

feat: introduced max reasoning effort tier across model and rpc systems

2026-07-10 13:39:42 +02:00

meta-provider.test.ts

feat: add native Meta Model API provider

2026-07-23 14:15:40 +00:00

minimax-code-login.test.ts

fix(ai): default MiniMax providers to M3

2026-06-10 08:31:30 +02:00

minimax-token-plan-usage.test.ts

fix(ai): honour the window status the endpoint reports as exhausted

2026-07-26 02:30:00 -03:00

mock-provider.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

model-cache.test.ts

fix(catalog): restored headers omitted from model cache

2026-07-17 03:27:12 +00:00

models-cost.test.ts

fix(catalog): bill cache-write tokens the ttl breakdown omits

2026-07-28 10:58:58 +02:00

models-json-no-local-endpoints.test.ts

feat: consolidated and automate changelog management

2026-06-27 03:08:52 +02:00

nanogpt-login.test.ts

feat: added injectable fetch transport across AI and coding network flows

2026-06-09 04:51:17 +02:00

novita-login.test.ts

fix(providers): fixed novita key validation for restricted accounts

2026-07-29 23:08:52 +02:00

null-max-tokens-fallback.test.ts

fix(ai): added finite output-token fallback when model maxTokens is unknown

2026-06-13 15:52:06 +02:00

oauth-barrel-import.test.ts

test(ai,catalog): give spawn-based lazy tests explicit timeouts

2026-07-29 23:09:06 +02:00

oauth-deepseek.test.ts

feat: added injectable fetch transport across AI and coding network flows

2026-06-09 04:51:17 +02:00

oauth-definitive-failure.test.ts

feat(ai): implemented idempotent credential refresh with CAS logic

2026-06-25 04:31:58 +02:00

oauth-device-code.test.ts

fix(plugins): loaded legacy extensions safely

2026-06-26 00:06:25 +00:00

oauth.ts

refactor(ai): consolidated provider config into single-source registry

2026-06-08 18:48:43 +02:00

ollama-cloud-login.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

ollama-no-user-turn.test.ts

fix(ai): guarantee ollama user turn and surface done_reason load

2026-08-03 07:33:56 +00:00

ollama-reasoning-effort-backfill.test.ts

feat: introduced max reasoning effort tier across model and rpc systems

2026-07-10 13:39:42 +02:00

ollama-thinking-disable.test.ts

fix(ai): retried empty ollama completions

2026-07-06 02:31:23 +00:00

ollama-tool-call-json-parse-error.test.ts

fix(ai): stripped transient flag for llama.cpp tool-call parse errors

2026-06-30 13:34:33 +00:00

openai-codex-include.test.ts

fix: resolved AI/OAuth helper duplication via shared modules

2026-04-23 21:02:14 +02:00

openai-codex-reset.test.ts

feat(coding-agent): rewrote codex saved-reset trigger algorithm

2026-07-31 20:39:12 +02:00

openai-codex-responses-lite.test.ts

chore(changelog): normalized unreleased entries after merges

2026-08-05 02:32:48 +02:00

openai-codex-stream.test.ts

chore: cleanup

2026-08-05 03:07:16 +02:00

openai-codex-usage.test.ts

fix(ai): respected explicit Codex usage allowance

2026-08-04 10:56:57 +00:00

openai-codex-zstd.test.ts

style: format codex zstd test

2026-07-29 23:10:58 +02:00

openai-codex.test.ts

feat: introduced max reasoning effort tier across model and rpc systems

2026-07-10 13:39:42 +02:00

openai-compat-policy.test.ts

refactor(ai): removed reasoning suppression prompt

2026-07-02 05:36:37 +02:00

openai-completions-cache-affinity.test.ts

fix(openai): keep implicit cache boundaries automatic

2026-07-23 17:08:40 -04:00

openai-completions-compat.test.ts

test(catalog): cover DeepSeek policy without bundled models

2026-08-02 20:55:16 +02:00

openai-completions-disable-reasoning.test.ts

fix: resolved Fireworks Qwen models to openai thinking format

2026-06-20 09:25:36 +02:00

openai-completions-error-finish-reason.test.ts

fix(ai): mapped bare finish_reason error to retryable error message

2026-06-12 22:27:04 +02:00

openai-completions-progress-chunk.test.ts

fix(ai): disabled local first-event watchdogs

2026-07-24 15:19:14 +00:00

openai-completions-reasoning-disable-dialects.test.ts

feat(ai): consolidated OpenAI-family streaming and add OpenRouter API support

2026-06-17 21:36:48 +02:00

openai-completions-tool-result-images.test.ts

fix(coding-agent): scope computer capture limits

2026-07-26 02:53:01 -07:00

openai-completions-upstream-provider.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

openai-computer-contract.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

openai-first-event-timeout.test.ts

fix(ai): preserved disabled first-event sentinel

2026-07-24 15:33:27 +00:00

openai-max-output-tokens-cap.test.ts

feat: add native Meta Model API provider

2026-07-23 14:15:40 +00:00

openai-output-token-policy.test.ts

feat(gateway): map Vercel automatic prompt caching

2026-07-23 15:18:37 -04:00

openai-reasoning-effort-fallback.test.ts

feat: introduced max reasoning effort tier across model and rpc systems

2026-07-10 13:39:42 +02:00

openai-responses-cache-affinity.test.ts

fix(openai): reject unsupported Azure prompt caching

2026-07-23 17:56:19 -04:00

openai-responses-delta-input.test.ts

feat(ai): introduced assistant message phase tracking for stage transitions

2026-06-27 13:05:57 +02:00

openai-responses-developer-role.test.ts

refactor(ai): removed reasoning suppression prompt

2026-07-02 05:36:37 +02:00

openai-responses-empty-tool-result.test.ts

fix(ai): prevented inclusion of empty image placeholders in tool outputs

2026-07-11 05:47:55 +02:00

openai-responses-history-payload.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

openai-responses-omit-max-output-tokens.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

openai-responses-openrouter.test.ts

fix: fixed OpenRouter usage reconciliation and updated model catalog entries

2026-07-16 02:58:30 +02:00

openai-responses-orphan-repair.test.ts

fix(ai): respect Responses replay pair order

2026-07-26 16:33:24 +00:00

openai-responses-parallel-tool-calls.test.ts

test(ai): validated parallel tool call argument routing and chunk handling

2026-07-05 15:57:24 +02:00

openai-responses-parallel-tool-result-images.test.ts

fix(ai): ordered parallel tool outputs before images

2026-07-17 14:37:13 +00:00

openai-responses-sampling-params.test.ts

fix(ai): omitted unsupported reasoning sampling params

2026-07-15 19:30:53 +00:00

openai-responses-stateful.test.ts

fix(ai): recomputed cache breakpoints after history edits

2026-07-24 03:10:02 +02:00

openai-responses-stream-retry.test.ts

fix(ai): retry safe truncated responses streams

2026-07-17 16:48:02 -05:00

openai-responses-stream-terminal.test.ts

fix(ai): execute complete tools at output limit

2026-07-22 15:54:27 -05:00

openai-responses-system-prompt.test.ts

fix(ai): made credential-pattern redaction opt-in

2026-07-23 16:23:08 +02:00

openai-responses-tool-quarantine.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

openai-stream-terminal-close.test.ts

fix(ai): preserved zero cache placeholder usage

2026-07-10 04:24:45 +00:00

openai-tool-strict-mode.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

openrouter-login.test.ts

test: aligned full suite with merged sweep contracts

2026-07-08 16:52:28 +02:00

overflow-utils.test.ts

refactor: centralized AI error handling into a dedicated module

2026-06-27 10:44:13 +02:00

owned-stream-fabrication.test.ts

refactor: renamed grammar module to dialect with unified transcript rendering

2026-06-15 13:58:12 +02:00

owned-stream-native-toolcall.test.ts

feat(ai): enabled streaming tool call argument deltas and validated identifier formats

2026-07-01 00:54:01 +02:00

pi-native-client.test.ts

fix(cli): preserve cache benchmark affinity diagnostics

2026-07-23 15:40:07 -04:00

pre-response-timeout.test.ts

test: improved test null safety and type assertions

2026-07-11 11:31:08 +02:00

provider-fetch-override.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

provider-inflight.test.ts

Fix provider in-flight limiter races

2026-06-27 01:39:32 +02:00

provider-registry.test.ts

feat(ai): add Exa API key login

2026-07-25 19:08:02 -04:00

provider-response.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

proxy.test.ts

refactor(coding-agent): adjusted block resolution formatting and update tests

2026-07-31 20:55:37 +02:00

rate-limit-utils.test.ts

fix(ai): keep concurrency caps out of auth rotation

2026-08-05 02:32:36 +02:00

raw-sse-sdk-capture.test.ts

test: aligned ai and tui test contracts with recent stream and paint changes

2026-07-10 13:21:06 +02:00

register-builtins.test.ts

fix(ai,catalog): widen Bedrock stream-stall watchdog via model compat

2026-08-07 14:18:46 +08:00

remote-auth-store.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

request-debug.test.ts

fix(pr-3347): use worktree-local cleanup helper in ai tests

2026-06-27 02:04:51 +02:00

requires-effort.test.ts

feat(ai): clamped omitted and disabled reasoning on requiresEffort models

2026-06-12 08:21:06 +02:00

resolve-used-fraction.test.ts

test(ai): clarify zero-limit contract name

2026-06-28 18:40:13 +02:00

sakana-provider.test.ts

feat: added support for Sakana AI and Fugu provider

2026-06-22 06:55:18 +02:00

schema-arktype.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

schema-compatibility.test.ts

feat(ai): added unified normalize flow for Google/CCA schema handling

2026-05-16 19:04:09 +02:00

schema-dereference.test.ts

fix: dereference MCP tool schema $ref/$defs and suppress Ajv format warnings (#276)

2026-03-03 22:52:41 +01:00

schema-helpers.test.ts

fix(ai,coding-agent): graceful strict-schema degradation for patternProperties

2026-02-28 20:47:41 +01:00

schema-normalization.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

schema-strict-mode.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

schema-wire.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

service-tier-premium-requests.test.ts

fix(ai): stop forwarding auto service tier to codex

2026-08-03 15:23:35 +00:00

sse-debug.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

stream-auth-retry.test.ts

fix(ai): exclude concurrency caps from 403 rotation and test the Copilot gate

2026-08-05 02:32:35 +02:00

stream-markup-healing.test.ts

fix(ai): recognize indented markdown fences as fenced code

2026-07-16 08:14:38 +00:00

stream-timeout-defaults.test.ts

feat(ai): extended codex and stream timeout defaults to 300 seconds

2026-07-30 05:33:06 +02:00

stream.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

synthetic-login.test.ts

feat: added injectable fetch transport across AI and coding network flows

2026-06-09 04:51:17 +02:00

synthetic-usage.test.ts

test(ai): isolate Synthetic usage fixture

2026-07-23 09:32:51 +01:00

thinking-loop.test.ts

feat(ai/utils): increased the Gemini header runaway threshold

2026-06-30 20:27:52 +02:00

todo-op-whitespace.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

tokens.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

tool-argument-coercion.test.ts

refactor: migrated packages to internal utility modules and removed external dependencies

2026-08-05 13:39:09 +02:00

tool-call-loop-guard.test.ts

fix(agent): detected repeated tool-call turns

2026-07-01 02:47:08 +00:00

tool-call-without-result.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

tool-examples.test.ts

feat: implemented shared serialization utilities and standardized inventories

2026-08-03 05:06:49 +02:00

tool-inventory.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

total-tokens.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

transcript-render.test.ts

test(ai): cover renderDelimitedThinking literal-envelope unwrap

2026-06-16 14:31:23 +02:00

transform-messages-dedup.test.ts

feat: added build-time compatibility in ModelSpec/buildModel pipeline

2026-06-10 06:20:51 +02:00

transform-messages-malformed-tool-calls.test.ts

fix(images): preserve replay and transcript invariants

2026-07-16 03:48:59 +02:00

transform-messages-redact-sensitive.test.ts

fix(ai): made credential-pattern redaction opt-in

2026-07-23 16:23:08 +02:00

transform-messages-thinking-dialect.test.ts

feat(ai): improved reasoning block separation and validation

2026-07-05 16:25:12 +02:00

umans-login.test.ts

Address Umans provider review feedback

2026-06-15 04:33:05 -07:00

umans-usage.test.ts

fix(ai): preserve path-mounted base URLs in umans usage provider

2026-07-03 23:47:00 -07:00

unicode-surrogate.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

usage-attribution.test.ts

fix: fixed OpenRouter usage reconciliation and updated model catalog entries

2026-07-16 02:58:30 +02:00

usage-report-notes-schema.test.ts

feat: introduced omptype validation package and migrated workspace dependencies

2026-08-03 21:56:48 +02:00

utils-responses-id.test.ts

test: migrated test imports to package subpath exports

2026-06-08 19:03:55 +02:00

vercel-gateway-cache-controls.test.ts

fix(gateway): preserve Vercel Responses routing and TTL policy

2026-07-23 15:38:50 -04:00

wafer.live.ts

Remove Wafer Pass provider

2026-06-21 02:45:48 +02:00

xai-login.test.ts

feat(ai): add xAI API key login

2026-07-28 10:03:42 -07:00

xai-oauth-effort-strip.test.ts

fix(ai): preserved xai default reasoning behavior

2026-07-10 00:11:48 +00:00

xai-oauth-usage.test.ts

fix: reconciled merged tests and dead code with current main structure

2026-07-23 18:02:31 +02:00

xhigh.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

xiaomi-oauth.test.ts

fix(providers): switched xiaomi validation to mimo-v2.5

2026-07-01 05:53:42 +00:00

xiaomi-tp-login-integration.test.ts

refactor(catalog)!: split model catalog from pi-ai

2026-06-10 04:06:57 +02:00

zai-oauth.test.ts

feat(providers): browser OAuth sign-in for Z.AI (GLM Coding Plan)

2026-07-23 14:42:15 +03:00

zai-usage.test.ts

refactor(usage): renamed Z.AI feature quota label and tier identifier

2026-07-24 14:06:34 +02:00

zenmux-login.test.ts

feat: added injectable fetch transport across AI and coding network flows

2026-06-09 04:51:17 +02:00

zhipu-coding-plan-login.test.ts

style: bun run fix

2026-07-02 10:12:31 +00:00