$ git log --author="copyberry" --all --oneline
Every morning, RepoJournal's newsroom reads what shipped across the open-source projects developers depend on — and writes the wire. Your commits kept making the news. This is your clippings file: the stories where your work was the story.
200
stories filed
62
daily wires
2
newsrooms
$ grep -rn "author: copyberry" wires/ | sort -r
Cline desktop bundles shrink, goose unblocks its own clicks
openai-node and openai-python ship managed Responses WebSocket sessions
Codex enforces managed residency, openai-node fixes streaming and abort paths
- Keep MCP user interaction on the root thread openai/codex
Codex carries MCP App UI through history, openai-node fixes shared callback credentials
- Preserve MCP App UI metadata in tool-call events and history openai/codex
- Complete Windows sandbox uninstall cleanup openai/codex
Guardian reviewer startup moves into the pool, thread resume keeps collaboration mode
- Restore collaboration mode when resuming threads openai/codex
- Move Guardian reviewer startup into the pool openai/codex
- Route Guardian reviewers through ThreadManager for inline parents openai/codex
Codex reworks its TUI: scrollback, paste search, and session launching
- Preserve terminal scrollback when growing the TUI viewport openai/codex
- Open new sessions directly from the command center openai/codex
- Route pastes into the active history search query openai/codex
- Add worktree session creation to the agents overview openai/codex
- Extract Windows sandbox configuration preparation into a helper openai/codex
codex retires thread/rollback and adds plugin and access-program controls to the app-server API
- Remove the deprecated `thread/rollback` API openai/codex
- Expose disabled plugin settings in the app-server API openai/codex
- Expose available access programs in model discovery openai/codex
Agents API lands in both OpenAI SDKs, Python streaming errors normalized
Codex 0.154.0 adds GPT-6-Astra and experimental worktrees, Python SDK stops crashing on null output
GPT Image 2.5 lands across Python and Node SDKs
Realtime credentials isolated in openai-node
- Add live WebRTC voice conversations to the TUI openai/codex
- Group adjacent computer actions in the TUI openai/codex
- Make the managed app-server shutdown grace period configurable openai/codex
- Enable user verification for the bundled TUI on supported devices openai/codex
Codex TUI and openai-node fix stale state and timers
- Use server model defaults for fresh TUI startup openai/codex
- Allow guarded legacy resume with background migration enabled openai/codex
Codex voice-host ships real-time microphone audio; openai-node fixes streaming hangs
- Send processed microphone audio over RTP in voice-host openai/codex
- Add opt-in local audio devices to the voice helper openai/codex
- Support managed worktrees for interactive sessions and forks openai/codex
OpenAI ships retry fix, prompt cache diagnostics, sandbox security
- Preserve Windows managed deny reads in the sandbox CLI openai/codex
- Preserve root authorization context in Guardian reviews openai/codex
- Route Guardian approvals independently of async scoring openai/codex
- Expose managed WebMCP policy through the app server openai/codex
- Report managed filesystem policy in `codex doctor` openai/codex
macOS Codex sandbox hardened against terminal input injection
- Add GPT-6-Astra to Amazon Bedrock catalogs openai/codex
- Initialize the packaged GStreamer runtime in the voice host openai/codex
- Add an injectable attachment store to ThreadManager openai/codex
Windows sandbox provisioning goes live, SDKs gain canonical security models
- Recover deferred environments after provisioning failure openai/codex
- Expose loaded thread environments in app-server responses openai/codex
- Add experimental context management activation openai/codex
- Add an RMCP OAuth credential store adapter openai/codex
openai-python and openai-node release 3.7.0 and 7.9.0 with usage API updates
- Scope session MCP approvals to app account links openai/codex
- Bound Git root discovery for metadata enrichment openai/codex
Guardian REPL policy now sources from model metadata
- Restore agent navigation after TUI reconnects openai/codex
- Open the agents overview from an empty composer openai/codex
Guardian auth survives compaction, update_plan becomes opt-in
- Make the update_plan tool opt-in openai/codex
- Show actionable rate-limit banners in the TUI openai/codex
- Support package-style MCP server names openai/codex
Codex adds Vim search motions and MCP tool result hooks
- Add Vim search motions to the composer openai/codex
- Harden diagnostic report uploads openai/codex
- Restore thread cwd from owned settings snapshots openai/codex
Codex subagents inherit root service tier; openai-python releases 3.6.0
- Restrict cloud task credentials to trusted origins openai/codex
- Refresh MCP HTTP helper headers after authorization failures openai/codex
- Refresh runtimes for remote plugin state changes openai/codex
OpenAI's codex cuts locking, Guardian survives long sessions
- Roll over Guardian context before follow-up reviews openai/codex
- Align deny-read matching with executor path semantics openai/codex
OPENAI SHIPS SECURITY HARDENING AND OPTIONAL CALL IDS ACROSS SDKs
- Support standalone function outputs in turn routing openai/codex
- Add trusted access context for MCP metadata openai/codex
- Track window and fork positions in turn metadata openai/codex
X.509 Workload Identity Lands in openai-node
- Add thread ownership metadata for managed worktrees openai/codex
- Sanitize credentials from Git remote metadata openai/codex
- Add explicit remote executor connection refresh openai/codex
- Track image generation request IDs in analytics openai/codex
- Move relay helpers into exec-server test support openai/codex
Two security patches land across OpenAI SDKs
- Honor response budgets when reading skill resources openai/codex
- Harden project config when credential brokering is active openai/codex
- Honor disabled managed network configuration openai/codex
Codex Drops the 'Plan' Nudge
- Remove the Plan mode composer nudge openai/codex
- Distinguish Guardian review threads from subagents openai/codex
- Support `cua_repl` as a Node REPL-backed MCP server openai/codex
CODEX TIGHTENS THREAD CLASSIFICATION, NODE SDK ADDS X.509 TRANSPORT
- Allow exec callers to classify new threads openai/codex
- Add regression coverage for patch approval paging openai/codex
- Require content kinds for extension prompt fragments openai/codex
- Shut down resumed descendants when archiving thread trees openai/codex
- Classify contextual fragments with content kinds openai/codex
OpenAI SDKs Tighten Security and Polling Internals
- Expand browser and computer use requirements openai/codex
- Expose browser and computer-use requirements through app-server openai/codex
- Add Guardian internal session support openai/codex
- Add a response target picker to `/copy` openai/codex
OPENAI NODE SHIPS FOUR SECURITY FIXES, STREAMING SAFETY AND CREDENTIAL LEAKS PATCHED
- Update rmcp to 3.1.3 openai/codex
- Add hostname to the configurable TUI status line openai/codex
OpenAI SDKs ship Bedrock Runtime and credential injection fixes
- Harden TUI subagent navigation openai/codex
CODEX HARDENS SESSION STATE, NODE SDK TIGHTENS ZOD VALIDATION
- Scope TUI app directory state to the active context openai/codex
- Forward workload identity context during token exchange openai/codex
- Use paginated history for persistent exec threads openai/codex
- Keep active-turn model settings stable across updates openai/codex
- Show resume and fork status during TUI startup openai/codex
OPENAI-NODE SHIPS FOUR SECURITY FIXES OVERNIGHT
- Give Guardian V2 full tool action context openai/codex
- Add app-server support for reverting paginated threads openai/codex
- Preserve approval policies for auto-reviewed models openai/codex
- Add rustls fallback for local MCP HTTP requests openai/codex
CODEX HARDENS PLUGIN METRICS AND SESSION RESILIENCE WHILE NODE SDK FIXES STREAMING ERRORS
- Reconnect gRPC code-mode sessions after host restarts openai/codex
- Report the latest rejection from multiple network reviews openai/codex
- Collect plugin metrics from unified exec commands openai/codex
- Collect metrics from plugin shell commands openai/codex
PYTHON SDK GOES HTTPX2, NODE PATCHES PROTOTYPE POLLUTION, CODEX HARDENS STREAMING
CODEX TIGHTENS MEMORY WITH MOVE-ONLY SUBMISSIONS AS NODE SDK COMPLETES LINTING OVERHAUL
CODEX FIXES IMAGE TRANSPARENCY AND PATH RESOLUTION VULNERABILITIES
- Use native transparency in the imagegen skill openai/codex
- Add a feature flag to preserve apply_patch line endings openai/codex
- Add a line-ending preservation mode to `apply_patch` openai/codex
- Bound Cursor project path resolution openai/codex
- Forward install attempt IDs for remote plugins openai/codex
CODEX HARDENS IDENTITY AND AUDIT LAYER ACROSS PLUGIN, HOOK, AND PROCESS EXECUTION
- Add workload identity token exchange support openai/codex
- Prevent launch context from reaching child processes openai/codex
- Improve plugin install failure analytics openai/codex
- Generalize hook handler execution openai/codex
- Use step environments for Guardian approval reviews openai/codex
CODEX HARDENS STREAMS AND SANDBOXING AS SDK CLEANUP ACCELERATES
- Keep response streams alive through connection failures openai/codex
- Delegate remote process sandboxing to the executor openai/codex
- Interrupt active code-mode cells with their turn openai/codex
CODEX SHIPS ROLLOUT MIGRATION FRAMEWORK AND CONTEXT WINDOW TRACKING
- Add rollout migration tooling and background migration openai/codex
- Track context windows per agent openai/codex
- Send model routing hints to the Codex backend openai/codex
- Fix subagent MCP startup status settling openai/codex
CODEX HARDENS SKILL SELECTION AND SESSION LOOKUP, NODE SDK CLOSES LINTER GAPS
- Load host skill roots through the skills extension openai/codex
- Harden named session lookup in the TUI openai/codex
- Test remote environments reported ready before selection openai/codex
CODEX HARDENS SECURITY, SDKS MODERNIZE TOOLCHAINS
GPT-5.5 LANDS ACROSS NODE AND PYTHON SDKS, CODEX HARDENS MCP CONFORMANCE
- Harden Linux managed proxy helper lifecycles openai/codex
- Raise the host-owned Codex Apps catalog limit openai/codex
CODEX HARDENS LOGIN FLOW WITH ONBOARDING HINTS AND METADATA PROTECTION
- Expose onboarding hints in login completion notifications openai/codex
- Preserve SQLite thread metadata during goal mutations openai/codex
- Capture rollout budget units from response usage openai/codex
CODEX SHIPS AGENT PLUGIN OVERHAUL AND TUI PERFORMANCE FIXES
- Support portable Agent Plugins throughout installation openai/codex
- Avoid querying terminal size on every TUI redraw openai/codex
- Support two-stroke TUI key chords openai/codex
- Retain attempted tool metadata across prompts openai/codex
NODE AND PYTHON SDKS SHIP CONTENT PROVENANCE CHECKS, CODEX CONSOLIDATES ARCHITECTURE
- Consolidate MCP config editing in codex-core openai/codex
- Migrate Cursor-managed skills into Codex openai/codex
- Extract exec-server request dispatching openai/codex
CODEX CONSOLIDATES HTTP PLUMBING ACROSS CORE AND CLI
- Load cloud-managed servers in MCP CLI commands openai/codex
- Refine OpenAI docs skill source routing openai/codex
CODEX CLEANS UP ASYNC CHAOS WHILE SDK TWINS SHIP TRANSCRIPTION UPDATES
- Clean up side conversations in the background openai/codex
- Track parent turns for nested Codex requests openai/codex
- Update rusty_v8 to 150.4.0 openai/codex
- Use the shared HTTP client for TUI network checks openai/codex
- Use configured HTTP clients for all MCP OAuth requests openai/codex
OPENAI-NODE DROPS NODE 20, PYTHON SDK REQUIRES 3.10
- Route curated plugins by authentication mode openai/codex
- Prepare MCP and plugin recommendations concurrently openai/codex
CODEX FIXES THREAD REPLAY ERRORS AND TIGHTENS SHUTDOWN SEQUENCING
- Preserve terminal turn errors in replayed history openai/codex
- Track model and personality in world state openai/codex
- Add managed policy for in-app updates openai/codex
- Skip inactive TUI threads without pending user interaction openai/codex
CODEX SHIPS RESPONSIVE UI FIXES AND FRESHENS MENTION SEARCH STATE
- Make the keymap action menu responsive openai/codex
- Keep unified mention results fresh openai/codex
- Include item start times in completion events openai/codex
- Handle exec-server network policy requests in the client openai/codex
CODEX HARDENS MCP RUNTIME STABILITY ACROSS FIVE CRITICAL FIXES
- Make the Apps recovery exposure test deterministic openai/codex
- Enable resource reads for explicit executor skills openai/codex
- Refresh MCP runtimes across thread startup openai/codex
- Use current MCP authority for elicitation reviews openai/codex
- Refresh managed MCP requirements for active threads openai/codex
SDK RELEASES LOCK IN PROMPT CACHING AND SPEND CONTROLS
- Attribute command executions to trusted plugin scripts openai/codex
- Add trusted plugin script attribution openai/codex
- Align installed app duration metrics with the legacy baseline openai/codex
- Support incremental replay of updated thread items openai/codex
- Expose remote skill icon URLs through app server openai/codex
CODEX CONSOLIDATES THREAD STARTUP, NODE SDK ADDS REALTIME AND STANDARD SCHEMA SUPPORT
- Consolidate thread startup around `StartThreadOptions` openai/codex
- Centralize SQLite connection configuration openai/codex
- Simplify app-server integration test setup openai/codex
- Upgrade Bazel Rust and LLVM dependencies openai/codex
CODEX FIXES WINDOWS EXECUTION BUGS WHILE TIGHTENING SECURITY REVIEW GATES
- Fix Windows TUI navigation key handling openai/codex
- Terminate Windows process trees with job objects openai/codex
- Load paginated model context across rollout lineages openai/codex
CODEX TIGHTENS MEMORY HANDLING AND AUDIO PRESERVATION ACROSS THREADS
- Add configurable hook context spill limits openai/codex
- Ignore inherited ACEs when refreshing Windows write roots openai/codex
- Enable memories for paginated threads openai/codex
- Preserve audio across history and tool outputs openai/codex
CODEX FIXES RACE CONDITIONS AND STABILIZES STREAMING PIPELINE
- Avoid liveness races when starting side conversations openai/codex
- Use the Markdown collector as the streaming source of truth openai/codex
- Speed up TUI Markdown layout openai/codex
- Track TUI command completion separately from output openai/codex
CODEX HOTFIX CORRECTS GPT-5.6 CONTEXT WINDOWS AFTER BUNDLED INSTRUCTION REFRESH
- Seed realtime V3 sessions with initial text items openai/codex
- Add audio output support to dynamic tools and code mode openai/codex
- Support legacy views for paginated thread history openai/codex
- Avoid redundant TUI redraws while streaming openai/codex
CODEX LOCKS DOWN MULTI-AGENT THREADS WHILE NODE AND PYTHON SDKS PATCH SECURITY GAPS
- Make parent-owned sub-agent threads read-only in the TUI openai/codex
- Add SessionEnd hooks for thread teardown openai/codex
CODEX SHIPS PAGINATED HISTORY INHERITANCE FOR SUBAGENTS
- Preserve paginated history for spawned subagents openai/codex
- Avoid creating metadata paths in the Windows sandbox openai/codex
- Propagate deferred environment capability roots to MCP openai/codex
- Add Cursor support to setup import openai/codex
- Refresh host skill catalogs through world state openai/codex
CODEX TIGHTENS MCP PLUMBING WHILE NODE SDK SHIPS ASYNC STREAMING
- Serialize concurrent MCP stdin writes openai/codex
- Reuse MCP tool catalogs across sessions openai/codex
- Remove task messages from `list_agents` output openai/codex
- Preserve plugin install failure subtypes during imports openai/codex
CODEX FIXES GUARDIAN ALIGNMENT AND THREAD HISTORY DURABILITY
- Resume thread history projection from its SQLite checkpoint openai/codex
- Prepare external agent migration for source adapters openai/codex
- Broaden remote compaction model fallback openai/codex
- Use model catalog policies for Guardian auto review openai/codex
A page like this — but yours, daily.
These clippings exist because your work kept showing up in other projects' wires. RepoJournal can write the whole journal: your commits, PRs and releases, turned into a daily devlog you never have to write.
Free. We read your public contributions only - private repos are never summarised.