Uh oh!
There was an error while loading. Please reload this page.
move all tests under tests/ - #3
Merged
Merged
Conversation
parkerhancock
approved these changes
Oct 12, 2025
parkerhancock
left a comment
There was a problem hiding this comment.
Looks great—auto-compaction logic now drives the same SUMMARIZATION_PROMPT and the integration fixture mirrors the trigger marker. Full test suite passes locally.
DioNanos referenced
this pull request
in DioNanos/codex-termux
Nov 15, 2025
Critical bugfix - auto-update detection now working Fixes: - Version parser now handles -termux suffix correctly - Tag parser accepts both rust-v* and v*-termux formats - Added test coverage for Termux tag validation Technical details: - v0.58.0 upstream merge overwrote -termux suffix support - New upstream code rejected v*-termux tag format - Prevented version.json creation and update notifications Affected versions: v0.58.0 through v0.58.3 Termux patches (4 total): - Patch #1: Browser login fix (termux-open-url) - Patch #2: RAM optimizations (lto=false, codegen-units=16) - Patch #3: Auto-update URL (@mmmbuto/codex-cli-termux) - Patch #4: Auto-update detection (this release)
DioNanos referenced
this pull request
in DioNanos/codex-termux
Nov 27, 2025
Version alignment with upstream development. This is Patch #3 - Version Alignment.
4 tasks
DioNanos referenced
this pull request
in DioNanos/codex-termux
Feb 18, 2026
Patches applied: - Patch #1: Browser login fix (termux-open-url) - Patch #2: Compilation optimizations (lto=false, codegen-units=16) - Patch #3: Version alignment (0.104.0) - Patch #4: Auto-update URL (DioNanos/codex-termux) - Patch #5: Version parser (-termux suffix) - Patch #6: NPM package name (@mmmbuto/codex-cli-termux) - Patch #9: Auto-update execution (tui/src/main.rs) Updated files: - codex-rs/Cargo.toml: version 0.104.0 - codex-rs/tui/src/main.rs: add update_action execution - npm-package/package.json: 0.104.0-termux - patches/README.md: update validated version
This was referenced Mar 8, 2026
shaqayeq-oai added a commit
that referenced
this pull request
Mar 10, 2026
## Summary Foundation PR only (base for PR #3). This PR contains the SDK runtime foundation and generated artifacts: - pinned runtime binary in `sdk/python/bin/` (`codex` or `codex.exe` by platform) - single maintenance script: `sdk/python/scripts/update_sdk_artifacts.py` - generated protocol/types artifacts under: - `sdk/python/src/codex_app_server/generated/protocol_types.py` - `sdk/python/src/codex_app_server/generated/schema_types.py` - `sdk/python/src/codex_app_server/generated/v2_all/*` - generation-contract test wiring (`tests/test_contract_generation.py`) ## Release asset behavior `update_sdk_artifacts.py` now: - selects latest release by channel (`--channel stable|alpha`) - resolves the correct asset for current OS/arch - extracts platform binary (`codex` on macOS/Linux, `codex.exe` on Windows) - keeps runtime on single pinned binary source in `sdk/python/bin/` ## Scope boundary - ✅ PR #2 = binary + generation pipeline + generated types foundation - ❌ PR #2 does **not** include examples/integration logic polish (that is PR #3) ## Validation - Ran: `python scripts/update_sdk_artifacts.py --channel stable` - Regenerated and committed resulting generated artifacts - Local tests pass on branch
charley-oai pushed a commit
that referenced
this pull request
Mar 11, 2026
## Summary Foundation PR only (base for PR #3). This PR contains the SDK runtime foundation and generated artifacts: - pinned runtime binary in `sdk/python/bin/` (`codex` or `codex.exe` by platform) - single maintenance script: `sdk/python/scripts/update_sdk_artifacts.py` - generated protocol/types artifacts under: - `sdk/python/src/codex_app_server/generated/protocol_types.py` - `sdk/python/src/codex_app_server/generated/schema_types.py` - `sdk/python/src/codex_app_server/generated/v2_all/*` - generation-contract test wiring (`tests/test_contract_generation.py`) ## Release asset behavior `update_sdk_artifacts.py` now: - selects latest release by channel (`--channel stable|alpha`) - resolves the correct asset for current OS/arch - extracts platform binary (`codex` on macOS/Linux, `codex.exe` on Windows) - keeps runtime on single pinned binary source in `sdk/python/bin/` ## Scope boundary - ✅ PR #2 = binary + generation pipeline + generated types foundation - ❌ PR #2 does **not** include examples/integration logic polish (that is PR #3) ## Validation - Ran: `python scripts/update_sdk_artifacts.py --channel stable` - Regenerated and committed resulting generated artifacts - Local tests pass on branch
apanasenko-oai pushed a commit
that referenced
this pull request
Mar 11, 2026
## Summary Foundation PR only (base for PR #3). This PR contains the SDK runtime foundation and generated artifacts: - pinned runtime binary in `sdk/python/bin/` (`codex` or `codex.exe` by platform) - single maintenance script: `sdk/python/scripts/update_sdk_artifacts.py` - generated protocol/types artifacts under: - `sdk/python/src/codex_app_server/generated/protocol_types.py` - `sdk/python/src/codex_app_server/generated/schema_types.py` - `sdk/python/src/codex_app_server/generated/v2_all/*` - generation-contract test wiring (`tests/test_contract_generation.py`) ## Release asset behavior `update_sdk_artifacts.py` now: - selects latest release by channel (`--channel stable|alpha`) - resolves the correct asset for current OS/arch - extracts platform binary (`codex` on macOS/Linux, `codex.exe` on Windows) - keeps runtime on single pinned binary source in `sdk/python/bin/` ## Scope boundary - ✅ PR #2 = binary + generation pipeline + generated types foundation - ❌ PR #2 does **not** include examples/integration logic polish (that is PR #3) ## Validation - Ran: `python scripts/update_sdk_artifacts.py --channel stable` - Regenerated and committed resulting generated artifacts - Local tests pass on branch
charley-oai pushed a commit
that referenced
this pull request
Mar 11, 2026
## Summary Foundation PR only (base for PR #3). This PR contains the SDK runtime foundation and generated artifacts: - pinned runtime binary in `sdk/python/bin/` (`codex` or `codex.exe` by platform) - single maintenance script: `sdk/python/scripts/update_sdk_artifacts.py` - generated protocol/types artifacts under: - `sdk/python/src/codex_app_server/generated/protocol_types.py` - `sdk/python/src/codex_app_server/generated/schema_types.py` - `sdk/python/src/codex_app_server/generated/v2_all/*` - generation-contract test wiring (`tests/test_contract_generation.py`) ## Release asset behavior `update_sdk_artifacts.py` now: - selects latest release by channel (`--channel stable|alpha`) - resolves the correct asset for current OS/arch - extracts platform binary (`codex` on macOS/Linux, `codex.exe` on Windows) - keeps runtime on single pinned binary source in `sdk/python/bin/` ## Scope boundary - ✅ PR #2 = binary + generation pipeline + generated types foundation - ❌ PR #2 does **not** include examples/integration logic polish (that is PR #3) ## Validation - Ran: `python scripts/update_sdk_artifacts.py --channel stable` - Regenerated and committed resulting generated artifacts - Local tests pass on branch
viyatb-oai pushed a commit
that referenced
this pull request
Mar 11, 2026
## Summary Foundation PR only (base for PR #3). This PR contains the SDK runtime foundation and generated artifacts: - pinned runtime binary in `sdk/python/bin/` (`codex` or `codex.exe` by platform) - single maintenance script: `sdk/python/scripts/update_sdk_artifacts.py` - generated protocol/types artifacts under: - `sdk/python/src/codex_app_server/generated/protocol_types.py` - `sdk/python/src/codex_app_server/generated/schema_types.py` - `sdk/python/src/codex_app_server/generated/v2_all/*` - generation-contract test wiring (`tests/test_contract_generation.py`) ## Release asset behavior `update_sdk_artifacts.py` now: - selects latest release by channel (`--channel stable|alpha`) - resolves the correct asset for current OS/arch - extracts platform binary (`codex` on macOS/Linux, `codex.exe` on Windows) - keeps runtime on single pinned binary source in `sdk/python/bin/` ## Scope boundary - ✅ PR #2 = binary + generation pipeline + generated types foundation - ❌ PR #2 does **not** include examples/integration logic polish (that is PR #3) ## Validation - Ran: `python scripts/update_sdk_artifacts.py --channel stable` - Regenerated and committed resulting generated artifacts - Local tests pass on branch
owenlin0 pushed a commit
that referenced
this pull request
Mar 11, 2026
## Summary Foundation PR only (base for PR #3). This PR contains the SDK runtime foundation and generated artifacts: - pinned runtime binary in `sdk/python/bin/` (`codex` or `codex.exe` by platform) - single maintenance script: `sdk/python/scripts/update_sdk_artifacts.py` - generated protocol/types artifacts under: - `sdk/python/src/codex_app_server/generated/protocol_types.py` - `sdk/python/src/codex_app_server/generated/schema_types.py` - `sdk/python/src/codex_app_server/generated/v2_all/*` - generation-contract test wiring (`tests/test_contract_generation.py`) ## Release asset behavior `update_sdk_artifacts.py` now: - selects latest release by channel (`--channel stable|alpha`) - resolves the correct asset for current OS/arch - extracts platform binary (`codex` on macOS/Linux, `codex.exe` on Windows) - keeps runtime on single pinned binary source in `sdk/python/bin/` ## Scope boundary - ✅ PR #2 = binary + generation pipeline + generated types foundation - ❌ PR #2 does **not** include examples/integration logic polish (that is PR #3) ## Validation - Ran: `python scripts/update_sdk_artifacts.py --channel stable` - Regenerated and committed resulting generated artifacts - Local tests pass on branch
EthanLipnik pushed a commit
to EthanLipnik/codex
that referenced
this pull request
Mar 11, 2026
## Summary Foundation PR only (base for PR openai#3). This PR contains the SDK runtime foundation and generated artifacts: - pinned runtime binary in `sdk/python/bin/` (`codex` or `codex.exe` by platform) - single maintenance script: `sdk/python/scripts/update_sdk_artifacts.py` - generated protocol/types artifacts under: - `sdk/python/src/codex_app_server/generated/protocol_types.py` - `sdk/python/src/codex_app_server/generated/schema_types.py` - `sdk/python/src/codex_app_server/generated/v2_all/*` - generation-contract test wiring (`tests/test_contract_generation.py`) ## Release asset behavior `update_sdk_artifacts.py` now: - selects latest release by channel (`--channel stable|alpha`) - resolves the correct asset for current OS/arch - extracts platform binary (`codex` on macOS/Linux, `codex.exe` on Windows) - keeps runtime on single pinned binary source in `sdk/python/bin/` ## Scope boundary - ✅ PR openai#2 = binary + generation pipeline + generated types foundation - ❌ PR openai#2 does **not** include examples/integration logic polish (that is PR openai#3) ## Validation - Ran: `python scripts/update_sdk_artifacts.py --channel stable` - Regenerated and committed resulting generated artifacts - Local tests pass on branch
marmar9615-cloud pushed a commit
to marmar9615-cloud/codex
that referenced
this pull request
Apr 19, 2026
Adds a hosted Node.js gateway (Express + ws) and a single-page browser UI that bring the Codex agent into the browser. The server spawns a backend agent per WebSocket connection and bridges Codex's JSONL event stream (thread.started, turn.started/completed/failed, item.started/updated/ completed for agent_message, reasoning, command_execution, file_change, mcp_tool_call, web_search, todo_list) up to the UI. - `web/server.mjs`: cookie-keyed in-memory sessions (7-day TTL), per-session workdir, same-origin guard on credential endpoints + WS upgrade, SIGTERM→SIGKILL escalation for child processes. Backend selection: real `$CODEX_BIN exec --experimental-json` when API key is set, else bundled mock. Real backend forces approval_policy=never (the JSONL stream doesn't carry approvals — that's the JSON-RPC app-server, deferred as task openai#2). - `web/mock-codex.mjs`: emits realistic JSONL events for every item type + an out-of-band approval-request envelope so the UI's approval flow can be exercised without OpenAI keys. Stdin uses a JSON-line protocol so approvals can be interleaved with the prompt while the turn runs. - `web/public/{index.html,app.js,styles.css}`: vanilla ES module, no build step. Dark theme; sessions sidebar; streaming agent messages; tool-call cards; inline approval prompts; 13 slash commands with `/`-autocomplete; `@`-file autocomplete; settings/login modals; reconnecting WebSocket. - Workflow "Start application" reconfigured to `cd web && npm start` on port 5000. - E2E tested via Playwright (chat streaming, plan, approval flow with apply_patch, errors, slash commands, settings round-trip). - replit.md and web/README.md updated with the new architecture. Deviations from the original plan: ChatGPT/OAuth login and full interactive approvals on the real backend are deferred to follow-up tasks openai#2 and openai#3 since they require bridging the separate JSON-RPC app-server protocol — out of scope for the MVP gateway. MCP server management UI is deferred to follow-up openai#4.
victalejo added a commit
to victalejo/codrex
that referenced
this pull request
Apr 28, 2026
Phase 2.5, commits 3+4+5/7 (combined). Bundles `codrex login <provider>`, `codrex logout <provider>`, and `codrex login --list` into a single landing because they share the same provider registry, prompt helpers, and list-rendering logic. The user-proposed split into three commits would have left awkward stub states between them. # `codrex login <provider>` Three credential-supply paths, in priority order: 1. `--with-api-key` (stdin pipe) — recommended for CI/scripts. 2. `--api-key sk-...` (deprecated, hidden in --help) — emits a stderr warning explaining the key stays in shell history and pointing at --with-api-key. 3. Interactive `rpassword`-backed hidden prompt when stdin is a TTY. Other flags: - `--coding-plan` tags the saved credential as a Coding Plan key (writes `kind: "coding_plan"` into auth.json::providers). - `--test-connection` opt-in flag for non-interactive flows; the interactive flow asks `Test the connection? [Y/n]` automatically. - `--list` (mutually exclusive with all other flags via clap conflicts_with_all) prints the credential table. Test-connection details: - Hits `<base>/chat/completions` once with `max_tokens: 4` against the provider's default model and reports `(model, latency_ms)`. - On success, persists `last_verified` so `--list` can surface it. - On failure, **never deletes the saved credential** — the message tells the user to re-run `codrex login <provider> --test-connection` after fixing the issue. Transient network failures must not invalidate a freshly-saved key. Hidden prompt only fires when stdin is a TTY. Piping a key without the `--with-api-key` flag now exits with a clear error instead of hanging on a hidden prompt nobody will see. # `codrex logout <provider>` Surgical removal — only the named provider's entry is removed from auth.json::providers. OpenAI auth (and any other provider's credentials) are preserved verbatim. When the removal leaves auth.json empty, the file is deleted outright. Env-var hint (correction openai#2 in the plan): when the user runs `codrex logout minimax` while `MINIMAX_API_KEY` is still set in the shell, the post-logout message includes actionable guidance: unset MINIMAX_API_KEY # current shell # then remove from ~/.zshrc, ~/.bash_profile, or wherever you set it Calling `codrex logout` (no provider) keeps the upstream behavior of clearing the entire auth payload. # `codrex login --list` Renders a stable, sorted table: Provider Type Source Last verified minimax coding_plan /Users/.../auth.json 2026-04-27 openai api_key env: OPENAI_API_KEY — Source column distinguishes file path vs `keyring (macOS)` vs `keyring (libsecret)` vs `keyring (Windows credential manager)` vs `env: <VARNAME>`. Per the plan correction openai#3, **the API key itself is never printed** — not even truncated prefixes. Per correction openai#4, the source display always identifies the storage backend (file path vs keyring + platform). When no credentials exist anywhere (per correction openai#6), the output is an actionable onboarding hint instead of an empty table: No credentials configured. To get started: codrex login # OpenAI (interactive OAuth) codrex login minimax # MiniMax (paste API key) # Provider registry `KnownProvider::resolve(id)` is the single registration point. Adding Qwen / DeepSeek / GLM in a future phase becomes a 6-line match arm — the prompt, save, list, and logout paths are provider-agnostic. Wiring: - `cli/src/provider_login.rs` (new) — handlers + KnownProvider catalog. - `cli/src/lib.rs` re-exports `run_login_provider`, `run_provider_logout`, `run_login_list`, `ProviderLoginInput`. - `cli/src/main.rs` extends LoginCommand with `provider`, `--coding-plan`, `--test-connection`, `--list`. LogoutCommand gets an optional `provider` arg. - `cli/Cargo.toml` adds chrono, codex-minimax, reqwest, rpassword, codex-utils-home-dir. - `Cargo.toml` workspace adds rpassword 7.3. Verified live against the freshly-built binary: - `echo sk-fake | codrex login minimax --with-api-key --coding-plan` saves credentials with `kind: "coding_plan"`, reports `✓ Permissions verified: 0600`. - `codrex login --list` after the above shows minimax row only, source=`<auth.json path>`, key never printed. - `MINIMAX_API_KEY=sk codrex logout minimax` removes the auth.json entry, deletes the file when last credential, surfaces the unset-from-shell hint. - `codrex login --list` with only env vars set shows `Source: env: MINIMAX_API_KEY`, key never printed. - `codrex login bogus-provider --with-api-key` → `unknown provider 'bogus-provider' (known: minimax)`. - `codrex login --list minimax` → clap rejects with `the argument '--list' cannot be used with '[PROVIDER]'`. Cross-cutting integration tests + docs/auth.md land in the final two commits of this phase.
victalejo added a commit
to victalejo/codrex
that referenced
this pull request
Apr 28, 2026
…n logging Phase 3 commit 4 — replaces PlaceholderAuditor with PatternAuditor that evaluates every AcceptanceCriterion variant against a real DispatchOutcome and produces a structured AuditReport with per-criterion timing, details, and a final AuditDecision. Decision matrix and ordering (cheap-to-expensive, short-circuit): Order Criterion Cost On failure ----- -------------------- ------- ---------- 1 NoForbiddenPatterns micros Escalate 2 OutputMatches[*] micros Retry 3 TestsPass seconds Retry 4 Custom[*] seconds Retry NoForbiddenPatterns escalates (not retries) per the Phase 3 plan guardrail openai#3: "the model is unlikely to 'not use' a pattern if it didn't understand why" — so the spec needs a human to refine, not the model to retry. Trait refactor (commit 1's contract): - Auditor::audit() now returns AuditReport { decision, criterion_results }. - New types CriterionResult and AuditReport surface in the lib root. - LogStage::AuditCriterion added; emitted once per evaluated criterion BEFORE the aggregated audit row. Greppable in JSONL as `"stage":"audit_criterion"` for per-criterion failure dashboards. - run_orchestrate updated to log per-criterion + final decision, and to exit non-zero with a clear message on Retry/Escalate/Drop verdicts (real retry loop arrives in commit 5). CLI flags surfaced for manual E2E demo (replaced in commit 6 by the rules-driven classifier): - --forbidden <regex> (repeatable, regex pre-validated) - --require-output-match <regex> (repeatable, regex pre-validated) - --require-tests-cmd <cmd...> (whitespace-split, sets TestsPass) Process safety (Phase 3 plan guardrail openai#2): TestsPass and Custom spawn with a 30s hard timeout via tokio::time::timeout. Timeouts record a structured `error: "timeout"` failure and short-circuit to Retry. tokio::process::Command uses kill_on_drop(true) so child processes die when the auditor returns early. Live verification (Apr 27 17:11-17:18 UTC): Scenario 1 (--forbidden ABC --require-output-match ".+", spec passes) → verdict=ok, all 2 criteria passed Scenario 2 (--require-output-match '^STRICTLY_DONE$', mismatch) → verdict=retry, feedback cites the regex Scenario 3 (--require-tests-cmd <fake-test exits 1 with stderr>) → verdict=retry, feedback contains stderr snippet Scenario 4 (--forbidden '7' + prompt requiring 7) → verdict=escalate, blocking_issue shows matched pattern + snippet Side observation from scenario 4 attempts: MiniMax respects the forbidden_patterns list when it sees them in the system message (commit 3's render_system_message wires them in). In practice that means most forbidden-pattern tripwires get pre-empted by the model itself or surface as CLARIFY:; the auditor's escalate path catches the rest. Healthy belt+suspenders. Tests: - 9 new unit tests in audit.rs: placeholder_auditor_returns_ok_with_no_criterion_results pattern_auditor_passes_when_no_criteria_configured pattern_auditor_passes_when_all_criteria_pass forbidden_pattern_match_escalates_not_retries output_matches_failure_returns_retry auditor_short_circuits_on_first_failure tests_pass_failure_returns_retry_with_stderr_snippet timeout_records_structured_failure_and_short_circuits custom_script_failure_returns_retry criterion_results_record_duration_ms - 45 orchestrator tests passing total. - Workspace cargo build green.
ehrvs referenced
this pull request
in PixelatedMountainStream/codex
May 7, 2026
…ub catalog Closes the test-coverage Blocking findings from QA-002 (rust-qa-agent), test-auditor, rust-code-reviewer Important #2, and simplification F6: - core/src/session/tests.rs: integration test exercising swap_model_provider end-to-end. Asserts both halves of the model_provider_id <-> model_provider invariant move together (services.model_client provider id flips, state.session_configuration.provider flips, original_config_do_not_use.model_provider_id flips, auth_manager Arc unchanged). This is the test that would have caught the Critical bug fixed in 04cc95d. - core/src/config/edit_tests.rs: SetModelAndProvider TOML round-trip at profile scope, mirroring existing SetModel test shape. Plus a None-effort test that asserts a pre-existing effort value is preserved. - tui/src/model_catalog.rs: two unit tests for the try_list_models conditional stub — one with ollama in configured_provider_ids, one without — proving the code-reviewer Important #3 guard works. Drive-by cleanups (within Milestone-1 scope; bigger ones deferred per simplification F2/F3/F5): - tui/src/chatwidget.rs: deleted the model_selection_actions one-line shim (simplification F4 + rust-code-reviewer Important #4) and updated its single call site to pass preset.provider_id.clone() through directly. Eliminates a latent bug where a future local preset satisfying is_auto_model() would silently drop its provider id. - tui/src/app_command.rs: removed spurious #[allow(clippy::too_many_arguments)] from a 4-arg fn (QA-003). - core/src/session/mod.rs: added SAFETY doc on swap_model_provider explaining the available_during_task() == false invariant that makes the two-phase lock pattern safe (QA-004 documentation requirement). - core/src/session/session.rs: TODO(milestone-1-followup) at SessionSettingsUpdate.model_provider_id field to track the simplification F2 dead-field cleanup (high-churn, deferred).
This was referenced May 16, 2026
defims added a commit
to defims/lemurclaw
that referenced
this pull request
Jul 20, 2026
Covers the 4 deferred items from subproject 4's final review: - #1: TopBar cwd + model from ConversationState (reducer captures thread.cwd + model/rerouted; sendRequest-backed turn/start + thread/resume feed authoritative ThreadStartResponse/ThreadResumeResponse back via a synthetic responseMeta action) - openai#2: AgentPanel real sub-agent data (extract from collabAgentToolCall.agentsStates into state.subAgents) - openai#3: full styles.css var-ification (~50 hardcoded colors → theme variables) Key insight during research: ThreadStartResponse + ThreadResumeResponse both carry authoritative model + cwd via the Task 4.1 sendRequest channel, so #1b is clean (not partial as originally thought). Plan written across 4 Write/Edit calls to avoid the single-Write-too-long interruption that hit the first attempt (documented in plan §实现备注 openai#8).
defims added a commit
to defims/lemurclaw
that referenced
this pull request
Jul 20, 2026
Var-ifies remaining ~50 hardcoded colors. Adds status (--ok/--err/--warn/ --warn-bg/--warn-border), neutral-variant (--muted-strong/--muted-weak/ --border-strong/--border-weak/--code-bg/--cell-bg-alt), and cell-type (--exec-accent/--mcp-accent/--plan-accent/--plan-bg/--hook-accent/ --agent-final-bg) variables to each theme block. Dark/high-contrast palettes now fully differentiate.
KitPan added a commit
to KitPan/codex
that referenced
this pull request
Aug 5, 2026
- compose_task_prompt：登记表 escalation_note 注入任务模板（openai#11「on-request 依赖模型自觉，不教不会」）；只注入新会话，resume 历史已含不重复。 spec/plan 记录实际发射文本（审计真相）。 - plan_overrides：分派 overrides 抽为纯函数；reasoning 档经 model_reasoning_effort 下发（openai#11 三维分派），并写入会话模型亲和（openai#3）。 - observe：codex metadata fallback 警告识别为已知无害并降噪（openai#15）—— catalog JSON 方案否决：ModelInfo.base_instructions 必填且整体替换系统 提示词，静态目录会把某版提示词冻结进仓库，rebase 漂移雷。 - 预存 clippy 1.95 新 lint 清零（collapsible_match / trim_split_whitespace）。 33 lib + 10 e2e 全绿。登记表内容更新在外层 repo（qwen36-deep 深推理档条目）。 Authored by Claude Fable 5 (Claude Code).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.