Uh oh!
There was an error while loading. Please reload this page.
update summary to auto - #1
Merged
Merged
Conversation
rumple pushed a commit
to rumple/codex
that referenced
this pull request
Apr 24, 2025
shepting pushed a commit
to shepting/codex
that referenced
this pull request
Sep 16, 2025
Closed
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 21, 2025
Upstream changes (40+ commits): - New: codex-shell-tool-mcp MCP server - New: execpolicycheck CLI command - New: TUI animations toggle feature - New: Shell timeout increased to 1 hour - Refactor: execpolicy migration (execpolicy2 -> execpolicy) - Removed: tiktoken-rs, shell_command feature - Improved: FreeBSD portability, fuzzy search (8->20 results) Termux patches verified compatible: - #1 Browser login (termux-open-url) ✅ - #2 RAM optimizations (lto=false, codegen-units=16) ✅ - #8 Bash execution (sandbox, LD_*, shell detection) ✅ - #9 Auto-update execution ✅ Stats: 195 files changed, +5915 insertions, -2293 deletions 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
DioNanos referenced
this pull request
in DioNanos/codex-termux
Nov 27, 2025
Merged 61 commits from openai/codex upstream (Nov 21-27, 2025). ✅ ALL TERMUX PATCHES PRESERVED: - Patch #1: Browser login (termux-open-url) - VERIFIED ✅ - Patch #2: RAM optimization (lto=false) - VERIFIED ✅ - Patch #4: Auto-update URL (DioNanos/codex-termux) - VERIFIED ✅ - Patch #5: Version parser (-termux suffix) - VERIFIED ✅ - Patch #6: NPM package name (@mmmbuto/codex-cli-termux) - VERIFIED ✅ - Patch #9: Auto-update execution (npm install) - VERIFIED ✅ NEW UPSTREAM FEATURES: - Enterprise skip upgrade config (check_for_update_on_startup) - Unified exec pruning strategy - Custom env for unified exec process - Compaction events - Thread/turn ID in notifications - Config management app-server - Cargo audit workflow + cargo-deny - NetBSD build fixes - Windows sandbox enhancements - Documentation improvements UPSTREAM CHANGES: - 232 files changed - +11247 insertions, -7808 deletions - New codex-api/ and codex-client/ crates - Rollout session error improvements - Apply patch parsing fixes CONFLICTS RESOLVED: - README.md: Merged Testing section (Termux) + Documentation links (upstream) NEXT: Compilation + Category 13 test suite validation
DioNanos referenced
this pull request
in DioNanos/codex-termux
Jan 8, 2026
…8+ requirement - Upstream: OpenAI Codex rust-v0.79.0 - Termux patches validated: #1-#6, #8, #9 - Testing: 47/49 passed (2 skipped: WebSearch, Git) - Documentation: - Added GLM-4.7 quickstart guide (docs/GLM4.7-quickstart.md) - Updated README with GLM-4.7 Path 2 in Quickstart - Node.js requirement updated to >=18 (recommended >=22) - Web search safety warning added - Tagline updated to "Built from upstream" - Package: npm-package synced with main README - Binaries: codex (59M), codex-exec (34M), codex-tui (41M) verified
charley-oai added a commit
that referenced
this pull request
Jan 9, 2026
…put image above user message
charley-oai added a commit
that referenced
this pull request
Jan 10, 2026
Agent wouldn't "see" attached images and would instead try to use the view_file tool: <img width="1516" height="504" alt="image" src="https://github.com/user-attachments/assets/68a705bb-f962-4fc1-9087-e932a6859b12" /> In this PR, we wrap image content items in XML tags with the name of each image (now just a numbered name like `[Image #1]`), so that the model can understand inline image references (based on name). We also put the image content items above the user message which the model seems to prefer (maybe it's more used to definitions being before references). We also tweak the view_file tool description which seemed to help a bit Results on a simple eval set of images: Before <img width="980" height="310" alt="image" src="https://github.com/user-attachments/assets/ba838651-2565-4684-a12e-81a36641bf86" /> After <img width="918" height="322" alt="image" src="https://github.com/user-attachments/assets/10a81951-7ee6-415e-a27e-e7a3fd0aee6f" /> ```json [ { "id": "single_describe", "prompt": "Describe the attached image in one sentence.", "images": ["image_a.png"] }, { "id": "single_color", "prompt": "What is the dominant color in the image? Answer with a single color word.", "images": ["image_b.png"] }, { "id": "orientation_check", "prompt": "Is the image portrait or landscape? Answer in one sentence.", "images": ["image_c.png"] }, { "id": "detail_request", "prompt": "Look closely at the image and call out any small details you notice.", "images": ["image_d.png"] }, { "id": "two_images_compare", "prompt": "I attached two images. Are they the same or different? Briefly explain.", "images": ["image_a.png", "image_b.png"] }, { "id": "two_images_captions", "prompt": "Provide a short caption for each image (Image 1, Image 2).", "images": ["image_c.png", "image_d.png"] }, { "id": "multi_image_rank", "prompt": "Rank the attached images from most colorful to least colorful.", "images": ["image_a.png", "image_b.png", "image_c.png"] }, { "id": "multi_image_choice", "prompt": "Which image looks more vibrant? Answer with 'Image 1' or 'Image 2'.", "images": ["image_b.png", "image_d.png"] } ] ```
DioNanos referenced
this pull request
in DioNanos/codex-termux
Jan 10, 2026
- Updated workspace version to 0.80.0-termux - All Termux patches verified functional (#1-#6, #9 active) - Patch #8 no longer required (resolved upstream by PR openai#8951) - Process hardening removed from Codex CLI improves bash execution - Test suite: 49/49 passed (CODEX_TEST_REPORT_v0.80.0.md) - Upstream changes: * Thread fork endpoints (conversation branching) * Requirements/list API * Elevated sandbox onboarding NUX * Skills explicit invocation via V2 API * Metrics capabilities (otel/metrics module) * Removed process hardening from Codex CLI - Binaries compiled: codex (60M), codex-tui (42M), codex-exec (35M), codex-app-server (38M) - Updated patches/README.md for v0.80.0
DioNanos referenced
this pull request
in DioNanos/codex-termux
Jan 10, 2026
- Remove old test report (CODEX_TEST_REPORT_v0.79.0.md) - Update version references from 0.79.0 to 0.80.0 - Update upstream reference from rust-v0.79.0 to rust-v0.80.0 - Update test date from 2026-01-08 to 2026-01-10 - Update test results to 49/49 passed (0 skipped) - Update patch description: 8 patches (#1-#6, #9) - Patch #8 resolved upstream - Sync npm-package/README.md with root README.md
charley-oai added a commit
that referenced
this pull request
Jan 12, 2026
…put image above user message
SammySnake-d added a commit
to SammySnake-d/codex-teams
that referenced
this pull request
Jul 14, 2026
… (gap openai#1) Closes the last real capability gap in the user's story: a reviewer teammate can now DIRECTLY watch a working teammate's mid-turn tool milestones, instead of all monitoring funnelling through the lead. This is what makes 'teammate as reviewer of another agent' actually work. - team_coord: send_progress_to_lead generalized to send_progress — fans a kind:Progress milestone out to the LEAD plus every member whose subscriptions contains the working agent. Missing config degrades to lead-only (no regress). progress_recipients() dedupes and excludes the worker itself. Old name kept as a thin alias. - New team_watch / team_unwatch tools: the caller subscribes to / unsubscribes from a named teammate (writes the target into the caller's own TeamFileMember.subscriptions under the config lock). Available to teammate processes (a reviewer IS a teammate). Caller identity is derived, never trusted from the model. Registered across all 6 compile-forced points + dispatch. - tui protocol.rs milestone push now calls send_progress (the fan-out path). - Unit tests: progress_recipients (lead+subscriber, excludes self, missing-config lead-only), send_progress fan-out to real mailboxes, team_watch spec shape. Search hints avoid ordinary-agent trigger phrases (isolation test green). - Live-proven (scripts/teams_reviewer_live_test.sh, 3/3): a real reviewer teammate calls team_watch(worker) -> config subscription written; a real worker runs a shell tool -> its kind:progress milestone lands in BOTH the subscribed reviewer's inbox AND the lead's. 113 team unit tests pass. Bump TEAMS_VERSION -> team.11. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
SammySnake-d added a commit
to SammySnake-d/codex-teams
that referenced
this pull request
Jul 14, 2026
…boration` Some frontier models (e.g. gpt-5.6-sol via a custom/`responses` provider) NATIVELY reserve the `collaboration` function namespace with a fixed schema. Upstream sends its multi_agent_v2 sub-agent tools under exactly that namespace, so the provider rejects the whole tools array (`invalid_request: Function 'collaboration.spawn_agent' is reserved ...`). One poisoned tool rejects the entire request, so ALL tool use breaks on that model — reproducible with a plain `codex exec`, no `--enable teams`. team.10 passed only because it used gpt-5.5, which does not reserve the name. Change (see codex-rs/docs/FORK_PATCHES.md openai#1): - config/mod.rs: DEFAULT_MULTI_AGENT_V2_TOOL_NAMESPACE "collaboration" -> "codex_agents", and the shared usage-hint example kept in sync (functions.codex_agents.spawn_agent). - Updates the 7 tests whose local `const MULTI_AGENT_V2_NAMESPACE` asserted the planned namespace. Verified: - core spec_plan + config + subagent_notifications (codex_agents.spawn_agent spawn dispatch) + app-server turn_start namespace tests pass. (Pre-existing, unrelated: debug-build stack overflows in agent::control / tokio workers, and a flaky thread_resume timing test — none touch namespaces; use RUST_MIN_STACK to bypass.) - live scripts/teams_reserved_namespace_live_test.sh: real gpt-5.6-sol, NO --enable teams, NO config override -> 0 reserved-name rejection + a shell tool actually executes. Daily path fixed. - full a2a live on gpt-5.6-sol: create_team -> spawn real teammate -> teammate team_send REAL_A2A_OK -> lead inbox received. TEAMS_VERSION -> team.13 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
yeshenlougu added a commit
to yeshenlougu/codex
that referenced
this pull request
Jul 18, 2026
… menu, shortcuts - openai#1 Agent tool execution now gates through sandbox.RequestApproval() before every tool call. Safe ops auto-approve; dangerous ops trigger WebSocket → frontend modal flow. - openai#5 Auto-compress context when msg count > 40 to keep last 8 pairs. Prevents token limit errors in long conversations. - openai#8 Chat '+' button now opens a dropdown with file picker and slash command shortcuts (/spec, /steer, /execute). - openai#9 Global keyboard shortcuts wired: Ctrl+Alt+B → toggle right panel, Ctrl+Alt+S → sidetasks, Ctrl+Shift+G → review, Ctrl+T → browser, Ctrl+P → files. - openai#6 API key empty check: provider returns clear error when key missing.
defims referenced
this pull request
in defims/lemurclaw
Jul 20, 2026
…roject 4 final review) Final whole-subproject review found 2 Important + 1 Minor worth fixing before declaring subproject 4 done. Important #1: App.tsx — the composition root (modal host, Ctrl+T, Onboarding gate wrap, Esc safety-net) had zero test coverage. Added App.test.tsx with 8 integration tests: layout-under-gate, Ctrl+T / Cmd+T / plain-T (negative) opens transcript, model + theme buttons open their modals, Esc closes, and single-modal-open enforcement. Mocks useConversation + Onboarding + transport so App renders without a backend. To make the integration tests pass, stubbed Element.scrollIntoView in the test setup (jsdom doesn't implement it; Scrollback calls it on mount). This unblocks any future test that renders the full tree. Important openai#2: registerResponseHandler was called in useConversation's mount effect, but React runs child effects before parent effects — so Onboarding's sendRequest('getAuthStatus') could theoretically fire before the response router was installed. It worked only because lib.rs installs a no-op stub up front AND the backend response takes longer than the next microtask. Moved the router install to transport.ts module-top-level (idempotent, guarded for non-window envs) so it's live before any React effect. Kept registerResponseHandler() as a public idempotent shim; the useConversation call is now defensive, not load-bearing. Minor: Composer.tsx stale comment said slash/@-mention/file popups 'land in subproject 4' — they actually land in subproject 5. Fixed. Verification: 89/89 tests pass (was 81; +8 App integration), tsc clean, npm run build succeeds, cargo check clean for lemurclaw-gui.
defims referenced
this pull request
in defims/lemurclaw
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).
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
10 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.