# OpenRig

Self-hosted coordination for persistent teams of coding agents.

OpenRig is a command-line tool (`rig`, from the npm package `@openrig/cli`) and a local daemon that run your Claude Code and Codex agents in terminal sessions. There's no JavaScript or Python library to import; you use it from a terminal (`rig up`, `rig send`) or through an agent. Start with Get help or Help for agents.

- [Installing OpenRig as an agent](https://github.com/mvschwarz/openrig/blob/v0.6.10/skills/rigs/SKILL.md): the rigs skill covers install and the first view; add it with `npx skills add mvschwarz/openrig --skill rigs`

## Start here / Get help
- [Get help](https://openrig.dev/help.md)
- [Help for agents](https://openrig.dev/help/agents.md)
- [From your first pair to a team of your own](https://openrig.dev/guides/growing-your-team.md)
- [Run OpenRig across machines](https://openrig.dev/guides/multiple-machines.md)

## Pages
- [Overview](https://openrig.dev/docs.md)
- [What Is A Rig](https://openrig.dev/docs/what-is-a-rig.md)
- [Architecture](https://openrig.dev/docs/architecture.md)
- [CLI reference](https://openrig.dev/docs/cli.md)
- [MCP tools](https://openrig.dev/docs/mcp.md)
- [Configuration](https://openrig.dev/docs/configuration.md)
- [Troubleshooting](https://openrig.dev/docs/troubleshooting.md)
- [Faq](https://openrig.dev/docs/faq.md)
- [Specs Overview](https://openrig.dev/specs.md)
- [Taxonomy](https://openrig.dev/specs/taxonomy.md)
- [Arena](https://openrig.dev/arena.md)
- [Guides Overview](https://openrig.dev/guides.md)
- [Claude Vs Codex](https://openrig.dev/guides/claude-vs-codex.md)
- [Model Diversity](https://openrig.dev/guides/model-diversity.md)
- [Ha Pair](https://openrig.dev/guides/ha-pair.md)
- [Pod Vs Single](https://openrig.dev/guides/pod-vs-single.md)
- [Factory Topology](https://openrig.dev/guides/factory-topology.md)
- [Growing Organically](https://openrig.dev/guides/growing-organically.md)
- [Adversarial Review](https://openrig.dev/explore/adversarial-review.md)
- [Research Cluster](https://openrig.dev/explore/research-cluster.md)
- [Security Hardening](https://openrig.dev/explore/security-hardening.md)
- [Continuous Refactor](https://openrig.dev/explore/continuous-refactor.md)
- [Secrets Manager](https://openrig.dev/explore/secrets-manager.md)
- [Claude Managed Agents](https://openrig.dev/compare/claude-managed-agents.md)
- [Claude Agent Teams](https://openrig.dev/compare/claude-agent-teams.md)
- [Claude Subagents](https://openrig.dev/compare/claude-subagents.md)
- [Crewai](https://openrig.dev/compare/crewai.md)
- [N8n](https://openrig.dev/compare/n8n.md)
- [Openclaw](https://openrig.dev/compare/openclaw.md)
- [Paperclip](https://openrig.dev/compare/paperclip.md)
- [OpenRig](https://openrig.dev/index.md)
- [What is OpenRig](https://openrig.dev/what-is.md)
- [What's new](https://openrig.dev/whats-new.md)
- [Added: Tell setup which agent harnesses you use, and it installs and checks only those (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-setup-asks-about-what-use.md)
- [Added: A new setting keeps agent-to-agent escalations out of your Slack unless a task is addressed to a person (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-new-escalation-setting-queue-wake-human.md)
- [Added: A Slack decision button that doesn't count now gets a reply in the thread saying why (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-slack-decision-click-isn-t-recorded.md)
- [Added: New tests and repository checks (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-tests-repository-checks-scenario-rig-up.md)
- [Expanded: You can add to a rig by its name as well as its ID (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-add-takes-rig-s-name.md)
- [Changed: After upgrading the CLI, restart the daemon so it runs the new version (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-restart-daemon-after-upgrading-cli-run.md)
- [Changed: Capturing the last N lines of an agent's terminal now gives you exactly those lines (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-capture-lines-n-shows-last.md)
- [Changed: The short queue list now shows each task's summary, and leaves out empty fields instead of printing them blank (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-compact-rig-queue-list-rows-carry.md)
- [Changed: Launching an agent that has exited no longer says it's already running (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-launch-longer-calls-exited-agent.md)
- [Changed: Bringing a rig back now says so when an agent is missing part of its guidance, instead of reporting a full restore (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-up-rig-existing-reports-seat.md)
- [Changed: A task an agent has claimed now reads as claimed even when its wake-up wasn't confirmed (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-lastnudgeresult-reads-claimed-claimed-row-whose.md)
- [Changed: Chatroom history now shows the newest messages by default, so a long room no longer opens on its oldest (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-chatroom-history-returns-newest-messages.md)
- [Changed: A Codex agent that has hit its usage limit no longer reads as idle (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-codex-seat-showing-usage-limit-banner-reads.md)
- [Changed: Creating a task with verify now reports the wake-up and pickup as soon as they happen, instead of waiting 30 seconds (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-queue-create-verify-reports-agent.md)
- [Changed: Teams are guided to need fewer approvals, to say what a launch changes in your folder, and to tell you when an agent is waiting on you (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-teams-guided-ask-less-say-more.md)
- [Changed: No npm dependencies or install scripts changed, and the Node requirement is unchanged (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-npm-dependencies-added-removed-changed-node.md)
- [Changed: Opening terminals now uses Herdr, or plain tmux when Herdr is missing, never cmux (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-caller-daemon-s-terminal-api-names.md)
- [Changed: The docs were updated for 0.6.10 (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-what-changed-note-capability-map-changelog-writt.md)
- [Fixed: On WSL2, the first daemon start no longer refuses a free port after a slow check (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-wsl2-first-rig-daemon-start-after.md)
- [Fixed: A Claude agent that resumes under a new conversation ID now comes back confirmed instead of needing your attention (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-resumed-claude-seat-whose-session-id.md)
- [Fixed: Bringing a rig back now restores the OpenRig guidance that taking it down removed, and leaves your own text alone (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-up-existing-puts-back-openrig.md)
- [Fixed: A restore no longer takes a launcher's word for which conversation is running, so it can't report a false resume (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-restore-longer-trusts-launcher-s-word.md)
- [Fixed: Messages and queue wake-ups now reach Codex agents started through a launcher (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-codex-seats-behind-launcher-seat-whose.md)
- [Fixed: A message to an idle Claude agent is no longer held back by the faint suggested prompt in its input box (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-claude-s-faint-suggested-prompt-empty.md)
- [Fixed: An agent whose terminal session is gone now reads as detached, with the way to bring it back (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-seat-whose-tmux-session-confirmed-gone.md)
- [Fixed: The attach commands OpenRig prints now find your agents when the daemon runs on a non-default tmux server (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-attach-commands-name-right-tmux-server.md)
- [Fixed: Starting OpenRig now says why it skipped an existing kernel and how to bring it back (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-start-says-why-skipped-existing.md)
- [Fixed: A Codex agent's context reading no longer disappears after a quota-only usage record (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-codex-seat-keeps-last-context-reading.md)
- [Fixed: A fresh Codex launch keeps git paths that end in a space (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-fresh-codex-launch-keeps-git-metadata.md)
- [Fixed: OpenRig recognises a tmux session started inside a Herdr pane (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-tmux-started-inside-herdr-pane-recognised.md)
- [Fixed: A rig name with an underscore or percent sign in it now filters views exactly (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-view-show-lens-rig-name.md)
- [Fixed: Searching chat for a keyword like config.json or C++ now matches it exactly (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-ask-matches-chat-keywords-such.md)
- [Fixed: rig heartbeat reads queue files with Windows line endings, and picks its latest proof correctly on hosts not set to UTC (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-heartbeat-reads-queue-files-windows.md)
- [Fixed: Queue recovery lookups no longer scan every row of the queue (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-queue-s-recovery-lookups-longer-scan.md)
- [Fixed: A limit that isn't a number now gets a request error instead of a server error (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-limit-isn-t-number-returns-400.md)
- [Fixed: Usage commands print errors properly in JSON and read every sample in their window (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-usage-top-series-json-prints.md)
- [Fixed: Creating a workflow now prints what it made (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-workflow-instantiate-project-print-instance.md)
- [Fixed: Following a transcript keeps lines that end in a percent sign (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-transcript-tail-keeps-line-ending.md)
- [Fixed: A --json written after -- no longer turns an error into JSON (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10--json-after-longer-turns-error.md)
- [Fixed: rig env prints the daemon's error instead of a JavaScript one (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-env-prints-daemon-s-error.md)
- [Fixed: Reading a large file no longer cuts a character in half (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-file-read-over-1-mib-ends.md)
- [Fixed: Four small TUI fixes for keyboard input, emoji, paging and opening files at a heading (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-tui-counts-cr-lf-one-enter.md)
- [Fixed: Scope and progress files are updated only where they should be, and examples in them are no longer mistaken for real sections (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-scope-progress-progress-rows-go-under.md)
- [Fixed: Installing a package finds the end of a guidance block reliably (OpenRig 0.6.10)](https://openrig.dev/whats-new/v0-6-10-rig-package-install-looks-guidance-block.md)
- [Added: Slack status now shows how many incoming messages are waiting to be retried (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-rig-slack-status-shows-how-many.md)
- [Added: A refocus work packet is included, switched off, so nothing changes unless it's enabled (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-refocus-work-packet-release-off-default.md)
- [Changed: After upgrading the CLI, restart the daemon so it runs the new version (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-restart-daemon-after-upgrading-cli-run.md)
- [Changed: A message sent to another machine that may have arrived now says delivery is unconfirmed instead of failed, with the check to run before sending again (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-cross-host-rig-send-host-http-registered-host.md)
- [Changed: No npm dependencies or install scripts changed, and the Node requirement is unchanged (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-npm-dependencies-added-removed-changed-node.md)
- [Changed: The docs were updated for 0.6.9, and now say that launching an agent takes the rig's ID, not its name (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-what-changed-note-capability-map-changelog-writt.md)
- [Fixed: Approving work for a named workspace now records the approval in that workspace, not the daemon's own, and a workspace with no missions folder is refused (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-rig-scope-tier-approve-workspace-path.md)
- [Fixed: When you answer a Slack decision in your own words, your reply is kept on the task instead of being lost (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-typed-slack-reply-decision-structured-questions.md)
- [Fixed: A Slack alert for a task that has already closed is no longer posted when the gateway restarts (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-slack-alert-whose-first-post-failed.md)
- [Fixed: A Slack connection that stays open but stops delivering is now noticed and replaced, and status says whether delivery is confirmed (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-openrig-notices-slack-socket-mode-connection.md)
- [Fixed: A kernel whose agents recovered from a failed start now reads ready, with the earlier failure still noted (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-kernel-whose-first-start-failed-whose.md)
- [Fixed: Waiting in a chatroom keeps going through a slow or restarting daemon instead of ending early (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-long-rig-chatroom-wait-keeps-waiting.md)
- [Fixed: Claude Code installed from nixpkgs is recognised, so its agents read as running (OpenRig 0.6.9)](https://openrig.dev/whats-new/v0-6-9-claude-code-seats-installed-nixpkgs-recognised.md)
- [Added: In the TUI, j and k now move up and down the list when the command line is empty (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-empty-tui-command-line-j-k.md)
- [Added: Preview what relaunching a single agent would do, including what happens to the others, before anything changes (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-rig-launch-rig-seat-plan-previews.md)
- [Added: When Slack asks OpenRig to slow down for a few seconds, it now waits and tries the post once more instead of giving up right away (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-slack-rate-limits-post-asks-pause-10.md)
- [Added: A new guide explains how to name a team's pods and agents (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-topology-naming-reference-explains-how-name.md)
- [Changed: After upgrading the CLI, restart the daemon so it actually runs the new version (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-restart-daemon-after-upgrading-cli-run.md)
- [Changed: A few TUI shortcuts moved to capital letters, so agents typing commands like find or mission into the TUI get the whole command through (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-tui-shortcuts-could-swallow-typed-command.md)
- [Changed: Workflow validation now catches a step that could never finish, instead of letting the workflow stall there later (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-rig-workflow-validate-reports-step-cannot.md)
- [Changed: Skill setup commands now exit with an error when a skill you selected is skipped for uncommitted changes (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-rig-context-work-install-runtime-runtime-rig.md)
- [Changed: A launch plan now checks the daemon's version first and won't send the plan to a daemon too old to understand it, which would launch instead (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-rig-launch-plan-reads-daemon-s.md)
- [Changed: The check that keeps the Slack bot token on Slack's own hosts now sits in the private-file download itself (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-check-keeps-slack-bot-token-slack.md)
- [Changed: Building OpenRig from source no longer needs Unix shell tools for its final step (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-building-source-cli-tui-build-scripts.md)
- [Changed: When an agent is woken for a task, it now sees the task's summary right away (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-agent-woken-task-handed-reminder-shows.md)
- [Changed: The architecture and CLI reference docs are up to date for 0.6.7 (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-architecture-cli-reference-docs-brought-up.md)
- [Changed: The README, guides and architecture docs are up to date for 0.6.8 (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-readme-reference-guides-architecture-docs-brough.md)
- [Fixed: Editing one skill in the shared skill catalog no longer stops every agent's skills (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-one-skill-being-edited-managed-skill.md)
- [Fixed: If a launch plan's answer shows the daemon actually acted, the command stops with an error and tells you how to check what happened (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-if-answer-rig-launch-plan-isn.md)
- [Fixed: When a person replies inside a Slack thread OpenRig started, an agent's answer now lands in that same thread (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-person-replies-inside-slack-thread-openrig.md)
- [Fixed: An agent that goes quiet after compacting is still restored, instead of waiting until something wakes it (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-after-managed-compact-claude-agent-takes.md)
- [Fixed: Claude agents started with Remote Control now confirm who they are instead of staying unconfirmed (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-claude-agent-launched-remote-control-confirms-id.md)
- [Fixed: Shutting a team down still saves its recovery snapshot even if one agent's details can't be read (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-rig-down-still-saves-recovery-snapshot.md)
- [Fixed: Scope audit accepts slice folders numbered 100 and up (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-rig-scope-audit-accepts-slice-folders.md)
- [Fixed: Proof files that start with a byte-order mark keep it (OpenRig 0.6.8)](https://openrig.dev/whats-new/v0-6-8-rig-proof-add-file-keeps-leading.md)
- [Added: Your own Claude Code and Codex sessions get two OpenRig skills, kept current on every upgrade, including rigs, which walks an agent from installing OpenRig to a team doing the work (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-daemon-copies-two-skills-own-claude.md)
- [Added: When your agent finishes installing or starting OpenRig, it now opens the dashboard and the kernel's operator for you instead of waiting to be asked, and the same command brings them back later (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-after-agent-installs-starts-openrig-opens.md)
- [Added: Your agents open wherever your own agent is running: a herdr space, a new Terminal window or a Ghostty tab, or a new window from Claude Desktop, iTerm or VS Code on a Mac (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-view-opens-where-agent-runs.md)
- [Added: On a Linux desktop the view opens in GNOME Terminal, Konsole or xterm at a readable size (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-linux-desktop-view-opens-gnome-terminal.md)
- [Added: Install OpenRig with one command, or preview it first: the dry run prints the four steps and changes nothing (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-readme-getting-started-guide-help-guide-show.md)
- [Added: If permission prompts are slowing you down, the operator points you to the workshop team and tells you plainly what access it takes (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-if-permission-prompts-slowing-down-operator.md)
- [Added: Each team, or a single agent, can post to its own Slack channel, and rig slack verify checks the app is in each one (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-slack-channel-map-set-rig-or-seat-channel.md)
- [Added: React to an agent's Slack question with an emoji and the agent gets a task saying who reacted and how (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-emoji-reaction-any-message-openrig-posted.md)
- [Added: After an upgrade, your agents can read what changed and what to do about it, and the upgrade skill finishes by telling you what affects you (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-agent-upgraded-install-read-what-behaves.md)
- [Added: Agents can find the built-in starter and factory teams by topic from the start, because their leads publish who's on the team at first start (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-built-in-starter-factory-leads-publish-team.md)
- [Added: While you choose a first team, the operator shows you each team's graph in the shared TUI (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-while-presenting-first-team-choices-operator-sho.md)
- [Added: A mission can be checked against a ref other than main, and a ref that can't be resolved reads as INDETERMINATE rather than guessing (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-mission-declare-arrangement-source-integration-r.md)
- [Added: If a Codex agent needs a newer Codex, the operator explains it and offers to update it (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-operator-explains-codex-agent-needs-newer.md)
- [Added: Pi agents tell you at start whether they have a sign-in or provider key, and how to add one, and launch checks compare your Pi version with the tested one (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-launch-preflight-reports-pi-version-against.md)
- [Added: Pi agents using MiniMax models can get their API key passed through once you allow it (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-pi-agents-minimax-id-models-receive.md)
- [Added: If kernel agents fail at first start because your tmux build crashes, the getting-started guide shows how to check for it and recover (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-getting-started-guide-covers-kernel-agents-fail.md)
- [Added: To run OpenRig on Windows, use WSL2: the getting-started guide records one user's working setup and both ways a Pi agent can get credentials (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-wsl2-way-run-openrig-windows-getting-started.md)
- [Changed: After upgrading the CLI, restart the daemon so it actually runs the new version (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-restart-daemon-after-upgrading-cli-run.md)
- [Changed: rig setup now installs herdr for the view of your agents, by running herdr's own installer (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-setup-installs-herdr-default-macos.md)
- [Changed: On a Mac, setup offers Ghostty once and installs it only if you ask with --ghostty (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-mac-rig-setup-offers-ghostty-once.md)
- [Changed: rig terminal open now opens a new desktop window by default (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-terminal-open-view-opens-desktop.md)
- [Changed: Your own herdr config is used as it is (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-if-have-own-herdr-config-openrig.md)
- [Changed: rig doctor now tells you whether Claude Code and Codex are installed and signed in, the same way setup does (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-doctor-checks-claude-code-codex.md)
- [Changed: The advisor setting now points at an agent every install has, the kernel's advisor, and writing a restore packet without flags targets the agent's own rig (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-agents-advisor-session-defaults-kernel-s.md)
- [Changed: A bundle you create, or install from a GitHub link, is named after the rig it contains instead of a generic name (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-bundle-create-rig-up-github.md)
- [Changed: A periodic reminder now waits one full interval before it first fires, instead of firing as soon as you set it up (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-periodic-reminder-first-fires-one-interval.md)
- [Changed: Package commands resolve relative paths from the folder you run them in, not from the daemon's folder (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-package-validate-plan-install-resolve.md)
- [Changed: Codex agents in your teams no longer stop at launch to offer a Codex update (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-managed-codex-launches-pass-c-check.md)
- [Changed: A Claude agent's activity hook now lives in OpenRig's own state folder instead of a script inside your project (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-claude-agent-s-next-launch-activity.md)
- [Changed: The kernel view fits your terminal: in a narrow one the operator gets the whole first tab, in a wide one the dashboard and the operator share it (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-default-saved-kernel-view-s-layout.md)
- [Changed: Over SSH, in CI or anywhere without a display, you're told why no window opened and given the exact command to run instead (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-window-open-over-ssh-ci-display.md)
- [Changed: The TUI's Open terminals action works the same way and doesn't need herdr (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-tui-open-terminals-action-uses-same.md)
- [Changed: Your coding agent's rigs skill now covers the whole way from install to the operator: when the install is done, how to pass the operator your goal, folder and branch, and how to bring the operator's questions back to you (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rigs-skill-carries-whole-route-install.md)
- [Changed: Ask the CLI, setup or onboarding how to see your agents or the welcome screen, and you're pointed to the one command that opens them (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-cli-help-setup-s-next-steps.md)
- [Changed: If your agent's own permission rules block a step while it installs OpenRig, the guides tell it to show you the refusal instead of reporting the install as failed (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-if-agent-installing-openrig-has-tool.md)
- [Changed: Claude Code agents in your teams can read help for the commands that start and stop teams without a permission prompt, and allowed commands match however they're written (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-claude-code-team-agents-help-lifecycle.md)
- [Changed: The kernel's operator can look around with routine commands, read the /rigs pages and run its first-boot checks without stopping for your approval each time (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-kernel-s-operator-runs-routine-inspection.md)
- [Changed: Setup's permission question now recommends the same default your teams use, and offers extra remembered allowances only if you want them (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-setup-s-permission-question-matches-team.md)
- [Changed: An agent's long question in Slack arrives as numbered replies in its thread, and if it still can't be posted, the agent is told instead of you never seeing it (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-question-too-long-one-slack-message.md)
- [Changed: The capability map your agents read to learn what OpenRig can do is up to date for 0.6.6 and 0.6.7 (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-capability-map-agents-read-learn-what.md)
- [Changed: Before a handover or compaction, agents are taught to write their recap with one command, so whoever picks up the work starts from it (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-handover-compaction-skills-teach-agents-write.md)
- [Changed: Before compacting, the daemon moves the agent's finished restore map into place itself, so the agent no longer has to run a rename that could stop at a permission prompt (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-daemon-itself-moves-finished-restore-map.md)
- [Changed: When rig bundle check finds a problem, it tells you which file it's in (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-bundle-check-names-file-each.md)
- [Changed: Adding a context pack from Git fetches only the current commit by default, and falls back to a full clone, with a warning, if the server won't allow that (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-context-add-git-retrieves-current.md)
- [Changed: To submit a team, the docs now point you to the openrig-registry repository (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-docs-name-openrig-registry-place-submit-team.md)
- [Changed: Reinstalling a stopped team replaces conflicting files only in the folder it was installed into, after backing them up (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-reinstalling-over-stopped-team-backs-up.md)
- [Changed: If you script against the daemon, a bundle install refused for a conflicting file returns HTTP 400 with a code, and other known refusals return 409 with theirs (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-bundle-install-refused-conflicting-file-returns.md)
- [Changed: Picking up a blocked task clears what it was blocked on, and marking it blocked again without saying on what reuses the previous blocker (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-claiming-blocked-task-clears-blocker-rig.md)
- [Changed: When a workflow step is sent back and redone, every step that depends on it runs again, even ones that had already passed (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-workflow-step-sent-back-completes-again.md)
- [Changed: Workflow commands use the workflow file in the folder you're in (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-workflow-validate-instantiate-run-use.md)
- [Changed: Scripts that tear down a rig on another host can tell when it went wrong: rig down --host exits 2 when the remote teardown reports errors (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-down-host-exits-2-remote.md)
- [Changed: Before stopping a team, rig down shows one line per agent with what it was last doing, so you can see what you're interrupting (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-human-mode-rig-down-prints-one-line.md)
- [Changed: Waiting on a chatroom with a timeout now stops at the deadline with exit code 1, instead of printing a message that arrived too late (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-chatroom-wait-timeout-exits-1.md)
- [Changed: A rig spec with malformed pods, members or edges gets ordinary validation errors you can act on (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-spec-validate-reports-malformed-pods.md)
- [Changed: Messages from rig gateway human now say that OpenRig supports one configured human (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-gateway-human-messages-say-openrig.md)
- [Changed: If a running agent's terminal fails its identity check, OpenRig says it needs attention and rig restore-check tells you what to do (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-running-agent-whose-current-pane-fails.md)
- [Changed: To stop everything, the operator and the getting-started guide give you the order: each rig, the kernel last, then the daemon (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-stop-everything-operator-getting-started-guide-g.md)
- [Changed: Help examples and guides no longer mention teams that were retired (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-help-examples-reference-guides-longer-use.md)
- [Changed: The README, reference guides, architecture docs and changelog were brought up to date for 0.6.6 and then 0.6.7 (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-reference-guides-brought-0-6-6.md)
- [Fixed: The status-line command OpenRig writes for a Claude agent now quotes its paths (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-claude-agent-s-next-launch-openrig.md)
- [Fixed: Opening the same herdr view again takes you back to the workspace that's already open, and your existing windows are never resized (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-opening-same-herdr-view-again-reuses.md)
- [Fixed: A message or watchdog reminder can no longer answer a question that's waiting for you (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-message-watchdog-reminder-longer-answers-questio.md)
- [Fixed: After a daemon restart, an agent waiting for your approval still shows as needing attention, so it isn't missed (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-after-daemon-restart-agent-waiting-approval.md)
- [Fixed: If you reply in a Slack thread and also send the reply to the channel, the agent that asked still gets it (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-slack-reply-sent-also-send-channel.md)
- [Fixed: When Slack's connection stalls while reconnecting, the daemon keeps trying instead of going quiet (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-if-slack-s-connection-stalls-while.md)
- [Fixed: You can hand the same role over to a fresh agent as often as you need, and running a handover again recovers one that stopped halfway (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-seat-handover-seat-works-same.md)
- [Fixed: An agent taking over from a Codex agent now sees its predecessor's messages in the recap (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-agent-taking-over-codex-predecessor-sees.md)
- [Fixed: After Claude compacts a long-running agent's context, the agent still gets its refocus context instead of losing it to a prompt, and repeated failures raise one issue you can find (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-after-managed-claude-compaction-agent-s.md)
- [Fixed: Forcing a bundle install past its version check no longer skips the bundle's safety check (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-bundle-install-skip-version-check-force-stil.md)
- [Fixed: A waiting task's timer shows when it will really go off, never as a date in 1970 (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-waiting-task-s-timer-shows-will.md)
- [Fixed: A timer length OpenRig can't read, such as 7d, now gets an error instead of silently doing nothing (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-invalid-wake-after-such-7d-prints-error.md)
- [Fixed: Resuming a workflow at a failure ID that doesn't match is refused, instead of rerunning a different failure (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-workflow-resume-occurrence-id-matches.md)
- [Fixed: Running or watching a workflow that was aborted now exits with code 3 instead of waiting forever (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-workflow-run-watch-exit-3.md)
- [Fixed: Asking for chatroom history since a time it can't read now fails, instead of telling you there are no messages (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-chatroom-history-since-value-t.md)
- [Fixed: rig skill audit passes on a clean install, and its JSON output gains new fields (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-skill-audit-passes-clean-install.md)
- [Fixed: Reading one section of a context file by its address works with inline code and empty headings (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-context-get-pack-file-section.md)
- [Fixed: The Codex plugin files OpenRig generates in your repository are now covered by its block in .git/info/exclude, so Git ignores them (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-files-openrig-generates-under-codex-plugins.md)
- [Fixed: A running Claude agent launched with inline permission settings is no longer mistaken for one that needs attention (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-openrig-reads-claude-agent-s-session.md)
- [Fixed: When OpenRig types a command to resume a Claude conversation, it quotes the resume token, so the shell can't misread it (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-resume-token-quoted-openrig-types-resume.md)
- [Fixed: Guidance text with $$ in it keeps its exact wording when OpenRig updates its managed sections (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-guidance-text-containing-similar-stays-written.md)
- [Fixed: If Codex shows a sign-in or version notice at the very end of a restore, the agent is flagged for attention with the last lines of its screen, and a late "no saved session" starts fresh (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-sign-in-incompatible-client-notice-appears-last.md)
- [Fixed: OpenRig finds the conversation of a Codex agent whose home folder path contains spaces (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-openrig-finds-codex-agent-s-conversation.md)
- [Fixed: A team you started fresh and then shut down cleanly comes back with rig up, instead of being refused for lack of a snapshot (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-up-restores-agent-launched-fresh.md)
- [Fixed: If a plugin skill's source can't be used any more, the agent launches with its existing copy and a warning instead of failing (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-plugin-skill-copy-whose-remembered-source.md)
- [Fixed: Names with a % in them work for agents, pods and agent images (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-agent-node-pod-agent-image-names-containing.md)
- [Fixed: OpenRig handles tmux session names with | or leading and trailing spaces, and reads each agent's launch settings from its own session only (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-discovery-capture-keep-tmux-session-names.md)
- [Fixed: Opening a large view gets up to 45 seconds to finish, instead of reporting a timeout after 5 (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-terminal-open-waits-up-45.md)
- [Fixed: If you use the web UI, a terminal view that fails to attach closes that viewer and leaves the daemon running (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-web-ui-terminal-view-fails-attach.md)
- [Fixed: Watching a chatroom no longer jumbles messages at the switch from history to live (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-chatroom-watch-keeps-messages-order.md)
- [Fixed: rig usage top keeps agents that reused a name apart (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-usage-top-longer-mixes-up.md)
- [Fixed: Usage series filtered by start and end times compare the actual times, not the text of the timestamps (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-usage-series-since-until-compare.md)
- [Fixed: Reading a transcript's tail or searching it while it's being updated gives you a consistent copy (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-transcript-tail-grep-read-consistent.md)
- [Fixed: Watchdog history keeps the right entries when several share a timestamp (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-watchdog-history-keeps-right-entries-tied.md)
- [Fixed: A very long snapshot history no longer breaks the cleanup of old snapshots (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-very-long-snapshot-history-longer-breaks.md)
- [Fixed: Moving, shipping or closing a piece of work with rig scope copes with unusual folder paths, and a failed attempt leaves the mission unchanged (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-scope-slice-move-ship-close.md)
- [Fixed: Creating a mission or a piece of work copies your text exactly, and updating its details keeps your keys, comments and quoting (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-rig-scope-mission-create-slice-create.md)
- [Fixed: Naming a piece of work for a mission finds the right one, even when an unrelated folder has the same name (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-bare-name-mission-resolves-right-piece.md)
- [Fixed: A project declared at the root folder no longer ties with a more specific project when OpenRig works out which project your folder belongs to (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-project-declared-longer-ties-deeper-project.md)
- [Fixed: In the TUI, a rig, host, pod or agent address now works from any page (OpenRig 0.6.7)](https://openrig.dev/whats-new/v0-6-7-tui-rig-host-pod-agent-address.md)
- [Added: If the daemon loses its Slack connection for a while, top-level channel messages posted during the gap still reach your agents, late and marked as recovered, instead of being missed (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-top-level-slack-messages-posted-while-daemon.md)
- [Added: After install, the first agent you meet is the kernel's operator: it asks what you want to build and recommends one of three teams (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-after-install-openrig-kernel-s-operator.md)
- [Added: You don't need every provider a team was designed for: the operator adapts the team to the accounts you have, asks before launching, and passes your goal to the team's lead so you only explain it once (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-operator-fits-team-pick-providers-have.md)
- [Added: When you want ongoing work, the operator keeps track of it in OpenRig's workspace, not in your repository (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-continuing-work-operator-records-one-mission.md)
- [Added: Whether setup finishes ready, incomplete or as a dry run, it ends by telling you how to reach the operator (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-setup-prints-how-reach-operator.md)
- [Added: After install, the agent that installed OpenRig opens the kernel's dashboard, advisor and operator in a new terminal space, so you can see them without disturbing its own session (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-after-installing-agent-opens-new-terminal.md)
- [Added: One command opens the kernel's TUI, advisor and operator side by side on any kernel, Claude-only, Codex-only or mixed, with nothing to set up first (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-terminal-open-saved-kernel-works.md)
- [Added: Without herdr or cmux, you still get the exact tmux or ssh command to reach each agent, and the shared TUI is labelled as the dashboard, not the place to talk to agents (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-without-herdr-cmux-setup-tui-preview.md)
- [Added: Open a team's spec in the TUI to see how its agents fit together as a graph, then launch it from there: it asks for a working folder and runs the real rig up (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-spec-opened-tui-starts-graph.md)
- [Added: A team can be shared as a link to a folder on GitHub and installed straight from it, pinned to one commit and with a package digest you can check (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-up-rig-bundle-create-inspect.md)
- [Added: A team can come in several ready-made configurations, such as which runtime each agent uses, and you choose one when you install it (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-bundle-declare-presets-configurations-yaml.md)
- [Added: Before you install a team, you can see what it will do on your machine: its models, whether permission prompts are off, the files and startup actions it brings, what it writes and which outside domains it uses (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-see-team-before-install.md)
- [Added: Install a team and point all its agents at your project folder in one step, separately from where the team itself is installed (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-bundle-install-cwd-dir-sets.md)
- [Added: Team authors can check a bundle against the authoring rules on their own machine before sharing it (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-bundle-check-folder-checks-bundle.md)
- [Added: When you package a team, it can carry a context pack kept outside the rig folder and the project the team works in, so whoever installs it gets both (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-bundle-create-context-pack-dir-carries.md)
- [Added: Installing a team you already have now asks what to do: use it, stop and replace it, or cancel (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-installing-team-already-exists-shows-what.md)
- [Added: Publishing your team on openrig.dev/rigs takes one pull request that adds a single file pointing at your repository (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-share-team-open-pull-request-openrig-world.md)
- [Added: If you build tools around team bundles, the v1 formats are now documented, with JSON Schemas to validate against (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-v1-bundle-formats-documented-docs-reference.md)
- [Added: Install the workshop from openrig.dev/rigs for a four-agent team that builds in any repository: its lead plans from your goal and routes work to a builder, a reviewer and QA, and nothing is published until you say so (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-workshop-openrig-s-team-building-software.md)
- [Added: Agents you've already set to full bypass can launch without stopping at Claude Code's and Codex's first-launch warnings, and the rig remembers that choice through restores and handovers (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6--non-interruptive-rig-up-rig-bundle-install.md)
- [Added: A team's author can declare in its spec that it launches without first-launch warnings, and a flag you pass on the command line still overrides it (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-spec-declare-non-interruptive-true.md)
- [Added: A new Claude Code agent stopped at Claude's bypass-permissions warning no longer loses its startup instructions (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-if-fresh-claude-code-agent-stops.md)
- [Added: You can run Claude Code agents in Claude's own auto permission mode by picking one built-in policy, and see each agent's effective mode and where it came from (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-new-built-in-policy-builtin-auto-launches.md)
- [Added: Agents can look up which specialist to bring in for a job, and why, from rosters you keep, across teams and machines (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-roster-list-show-find-read.md)
- [Added: Tools that collect OpenRig's history can read it a page at a time through the CLI, with any gaps marked, instead of reading OpenRig's database directly (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-telemetry-events-transitions-tenures-read.md)
- [Added: A project can name the world packs its agents should install, alongside the rest of its context (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-project-list-world-packs-under-install.md)
- [Added: If you work from the repository, an optional script checks for Node and npm, installs OpenRig and runs setup, stopping at the first failure (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-repository-has-optional-scripts-install-sh.md)
- [Added: Agents that are slow to start can be given up to ten minutes to become ready, at launch, restore and handover, instead of the default 30 seconds (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-runtime-readiness-timeout-seconds-1-600.md)
- [Added: Inventory timeouts are easier to diagnose: requests marked with a diagnostic header are traced in the daemon's slow-operation log (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-requests-carry-x-openrig-diagnostic-attempt-head.md)
- [Added: The project commands shadow-drain and shadow-stop now work from an ordinary shell, not only from inside an agent (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-project-shadow-drain-shadow-stop-take-actor.md)
- [Added: Any Claude Code or Codex session, even outside an OpenRig team, gets a way into OpenRig by adding one skill with a single npx command (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rigs-skill-gives-ordinary-claude-code.md)
- [Changed: After upgrading the CLI, restart the daemon so it actually runs the new version (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-restart-daemon-after-upgrading-cli-run.md)
- [Changed: If you start any of these three teams by name, in scripts or by habit, switch to the new name (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-three-specialist-teams-renamed-adversarial-revie.md)
- [Changed: Instructions that say rig up first-project keep working and now start the starter team, which pairs a Claude Code builder with a Codex reviewer (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-up-first-project-still-works-starts.md)
- [Changed: Seven older built-in teams were retired (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-conveyor-demo-implementation-pair-product-team-f.md)
- [Changed: Implementer and QA agents now start without the test-driven-development skill loaded (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-implementer-qa-agents-longer-load-test-driven-de.md)
- [Changed: Claude Code agents in your teams can run OpenRig commands, read the project and run its tests without a permission prompt, but still ask before starting, stopping or installing teams (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-claude-code-agents-teams-need-fewer.md)
- [Changed: Codex agents in your teams get the same relief: routine OpenRig, read and test commands ran without prompts in testing, still inside Codex's sandbox (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-codex-agents-teams-need-fewer-permission.md)
- [Changed: The kernel's operator is the agent that installs and runs teams for you, so kernel agents now launch with the broad permissions that job needs, and Codex runs there without a sandbox (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-agents-built-in-kernel-rig-launch-wider.md)
- [Changed: Two rigs with near-identical names no longer end up sharing Docker containers by accident, because each new rig with services gets its own Compose project (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-new-rig-services-gets-docker-compose.md)
- [Changed: If a new rig's services fail to boot or none of its agents launch, OpenRig takes down the containers it just created and never deletes their volumes (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-if-service-boot-every-agent-launch.md)
- [Changed: OpenRig no longer hides the guidance files it adds to your repository: a new CLAUDE.md or AGENTS.md shows in git status, and OpenRig prints the exact ignore line if you want it hidden (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-files-launch-creates-show-git-status.md)
- [Changed: The restore map a Claude agent writes before its context is compacted now sits in the agent's own launch folder, with a .gitignore that keeps it out of your commits (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-managed-claude-compaction-writes-restore-map.md)
- [Changed: Ask your coding agent to install OpenRig and it now hands your goal and project folder to the operator, and leaves the kernel running, instead of building the project itself (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-agent-installs-openrig-keeps-kernel-running.md)
- [Changed: When the operator suggests the workshop, it checks the latest pinned version instead of trusting a cached copy that may be older (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-operator-offers-workshop-reads-workshop-s.md)
- [Changed: When a team starts, you know where to go: watch it from the dashboard, and talk about the work in the lead's pane (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-team-views-say-dashboard-overview-lead.md)
- [Changed: Open the TUI and you're at your work right away, even before any team is running, rather than on a start page (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-tui-goes-straight-work-views-once.md)
- [Changed: Previewing the starter or factory team ends with the same three-team choice, so you can compare before you pick (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-specs-preview-starter-rig-specs.md)
- [Changed: Everything a bundled team brings, from skills and plugins to context packs, is put in place before its agents start, and anything that can't be routed is reported instead of silently dropped (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-skills-plugins-workflow-specs-context-packs.md)
- [Changed: Installing a team never changes a context pack you already have under the same name (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-installing-bundle-never-overwrites-merges-instal.md)
- [Changed: Bootstrapping from a .rigbundle file now uses the same install path as rig bundle install, so both ways of installing a team behave alike (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-bootstrap-file-rigbundle-goes-through.md)
- [Changed: You can change an agent's permissions from an ordinary shell, not only from inside an agent, and an incomplete request tells you exactly which part is missing (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-seat-set-permissions-takes-operator-address.md)
- [Changed: With several projects in one workspace, an agent loading its project context gets one picked by clear rules and is told how it was chosen (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-several-projects-declared-project-rig-context.md)
- [Changed: Handing work to another agent returns as soon as the task is saved, so you aren't kept waiting on delivery (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-queue-create-handoff-handoff-and-complete-re.md)
- [Changed: rig doctor no longer gives a blanket all-clear, so a clean run isn't mistaken for a working setup: it points you to its warnings and says it doesn't check provider logins or agent readiness (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-doctor-longer-ends-system-checks.md)
- [Changed: Setup shows you what to do next before it asks about permissions, and scripts reading its JSON output get the next steps too (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-setup-guided-next-steps-come.md)
- [Changed: When setup runs into problems, you get a list of what failed and how to fix each one (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-setup-finishes-problems-lists-each.md)
- [Changed: New Claude agents finish setting themselves up without anyone typing to them: after their startup text, OpenRig sends one short line asking each to run its startup check (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-after-delivering-claude-code-agent-s.md)
- [Changed: Startup text that wasn't confirmed as sent now shows as a warning in launches, imports, restore records and the TUI, so an agent that may have missed its instructions is easy to spot (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-startup-files-delivered-later-get-same.md)
- [Changed: Oh My Pi agents using Anthropic, OpenAI or LiteLLM models can send requests to a custom endpoint, such as a gateway, once you allow that provider's base-URL variable (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-oh-my-pi-agent-whose-model.md)
- [Changed: Starting without the kernel now tells you the operator won't be running either, and how to start it later (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-daemon-start-no-kernel-rig-start.md)
- [Changed: Setup offers to show you the kernel before you pick a team, and treats saying no, working over SSH or running headless as fine (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-setup-s-next-steps-offer-kernel.md)
- [Changed: Before you stop or restart a running team, the help tells you what that does to the work in progress (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-lifecycle-help-says-what-stopping-restarting.md)
- [Changed: The getting-started and agent help guides now cover what to do when an agent needs attention at startup, including Claude's bypass warning, slow-starting agents and the auto permission policy (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-getting-started-guide-agent-help-guide-explain.md)
- [Changed: Common install snags on Linux are now in the README and getting-started guide: a system Node.js that's too old, npm 11 skipping the install check, and cmux being optional and macOS-only (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-readme-getting-started-say-linux-distribution-s.md)
- [Changed: Adding context works even when the daemon is stopped, because the command starts it for you (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-context-add-starts-stopped-local.md)
- [Changed: Agents working on OpenRig's own code are pointed to the skill that explains how OpenRig is built (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-bundled-skills-index-sends-agents-changing.md)
- [Changed: If you read or contribute to OpenRig's code, its developer docs were refreshed throughout, with each claim checked against the code at the time (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-developer-documentation-refreshed-each-claim-che.md)
- [Changed: Contributors install exactly the dependency versions in the lockfile, so everyone builds against the same set (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-contributor-setup-uses-npm-ci.md)
- [Changed: OpenRig's contributor guidance now asks whether an agent with instructions could do a job before anyone writes code for it (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-contributor-skill-asks-whether-agent-instruction.md)
- [Redesigned: The built-in teams are now a shorter list with plainer names: starter (a builder and a reviewer) for a first team, factory (a lead and six other agents) for bigger work, and specialist teams such as code-review and research (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-built-in-teams-have-new-names.md)
- [Fixed: If you run OpenRig on Linux, the kernel's advisor no longer talks to you as if you're on a Mac (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-kernel-s-advisor-agent-longer-assumes.md)
- [Fixed: Every bundle now records which configuration it was built from and the OpenRig version that built it, so you can tell later where it came from (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-every-rig-bundle-create-records-configuration.md)
- [Fixed: Skills that come in a plugin, including OpenRig's own, are now copied to the folders Claude Code and Codex actually load skills from, which never happened before (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-agent-s-profile-selects-plugin-such.md)
- [Fixed: The attention view, the health readout and rig mode effective now use the projects you configured and pick a task's project the same way project context does, so they agree on which project a task belongs to (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-mode-effective-attention-view-health.md)
- [Fixed: Answering an agent's numbered menu from outside its terminal now sends the digit as a keypress rather than a paste, so the menu gets the choice you meant (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-send-dangerously-interact-types-answer-keyst.md)
- [Fixed: A setup dry run on Linux now reports the Homebrew and cmux steps as skipped, so the preview matches your system (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-linux-rig-setup-dry-run-reports-homebrew.md)
- [Fixed: If an agent is stuck at a provider's own prompt, such as Claude's trust dialog, OpenRig flags it for your attention instead of calling it ready, so you know someone has to answer it (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-agent-stopped-provider-s-own-prompt.md)
- [Fixed: OpenRig reads more Claude Code screens correctly: an agent with only update or limit warnings under an empty prompt counts as idle and gets its messages, and one that's visibly working isn't mistaken for idle (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-claude-code-pane-read-idle-update.md)
- [Fixed: When a Claude Code agent asks a question with a long list of choices, OpenRig now recognises that it's waiting on an answer (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-tall-claude-code-question-picker-recognised.md)
- [Fixed: Inspecting a bundle before you install it now applies the same archive safety checks install does, so an unsafe archive is caught at the first look (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-bundle-inspect-refuses-same-unsafe.md)
- [Fixed: An older-format bundle whose project settings point outside its own folder is now refused (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-legacy-bundle-whose-project-block-escapes.md)
- [Fixed: An agent can carry its own version of a skill: the launch uses it and warns that it differs from OpenRig's catalog, instead of failing (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-profile-selects-skill-own-agent-folder.md)
- [Fixed: A package can no longer use a skill or agent name to write files outside its install folder (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-installing-package-whose-manifest-gives-skill.md)
- [Fixed: A rig with a Pi agent no longer fails its pre-launch check just because the daemon can't see pi on its PATH, since the agent's own shell finds it at launch (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-pi-agent-longer-fails-preflight.md)
- [Fixed: If rig down can't stop an agent, that agent keeps its CLAUDE.md or AGENTS.md, so it isn't left running without its instructions (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-down-t-stop-agent-leaves.md)
- [Fixed: Stopping services through the daemon's API removes their data volumes only when the request explicitly asks for it (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-tearing-down-services-through-daemon-s.md)
- [Fixed: rig host pair always gives up eventually, even when given an infinite or oversized timeout (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-host-pair-timeout-caps-infinite.md)
- [Fixed: Claude agents launched through a shell wrapper named with a version number are now recognised by their executable path, so a restore problem flagged on them can be cleared (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-openrig-identify-claude-process-running-under.md)
- [Fixed: Claude agents brought back with rig up --existing keep reporting their activity to OpenRig, and a restore that can't set that up again says so (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-agent-brought-back-rig-up-existing.md)
- [Fixed: Forking a Claude agent now quotes the parent conversation's ID in the launch command, so the shell can't misread it (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-forking-claude-agent-quotes-parent-session.md)
- [Fixed: Codex agents no longer get stuck at launch, resume or fork when Codex 0.160 shows its newer update menu (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-codex-agents-longer-stall-launch-resume.md)
- [Fixed: Pi agents keep talking to the right daemon as the right agent, even when your shell startup files set conflicting values (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-pi-agents-keep-daemon-routing-identity.md)
- [Fixed: When a Pi agent can't find an API key, the error now tells you what to set (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-pi-agent-failing-api-key-found.md)
- [Fixed: When Oh My Pi quits during startup, you see how it exited and its last error output, so you can tell why (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-oh-my-pi-exits-during-startup.md)
- [Fixed: When rig whoami can't find the agent, the error names the daemon it reached, so you can tell if it's talking to the wrong install (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-whoami-rig-health-self-t.md)
- [Fixed: Marking a task as blocked is easier to get right: the help says you must name what it's waiting on, and naming a task this daemon doesn't have suggests waiting on something outside OpenRig instead (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-queue-block-help-says-on.md)
- [Fixed: The TUI's recent task activity takes less work to load when there's a long history behind it (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-query-behind-tui-s-recent-queue.md)
- [Fixed: OpenRig's saved Slack connection state stays intact if the daemon crashes while writing it (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-slack-gateway-state-survives-crash-middle.md)
- [Fixed: A context-pack folder created in the workspace after the daemon started is now found by sync, without restarting the daemon (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-context-sync-finds-workspace-context-pack.md)
- [Fixed: Context packs that include JavaScript or Python helper files now install from the command line (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-context-packs-mjs-py-helper-files.md)
- [Fixed: Pointing rig project experimental at a named pipe now gets a clear refusal instead of a hang (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-project-experimental-refuses-fifo-instead.md)
- [Fixed: rig capture now shows an empty agent screen as empty (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-capture-shows-empty-pane-empty.md)
- [Fixed: OpenRig's MCP server exits when its input closes, and keeps using the daemon address it found, including IPv6 ones (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-mcp-serve-shuts-down-input.md)
- [Fixed: Mission titles with colons, hash signs or quotes no longer break the mission's file (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-mission-title-quotes-writes-valid-frontmatter.md)
- [Fixed: rig env logs --tail 0 now prints no earlier lines, as you asked (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-env-logs-tail-0-prints.md)
- [Fixed: rig env status keeps working when a cached service record is damaged (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-env-status-survives-corrupt-cached.md)
- [Fixed: rig terminal open now uses the daemon's own tmux, and with herdr it reports a pane that closes at once as degraded instead of treating it as working (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-terminal-open-starts-tiles-daemon.md)
- [Fixed: Agent transcripts now read correctly where text was erased with several backspaces in a row (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-transcript-cleanup-applies-runs-backspaces-corre.md)
- [Fixed: Skills and commands no longer point you to skills OpenRig doesn't ship, and rig compact-plan names the compaction restore skill that does (OpenRig 0.6.6)](https://openrig.dev/whats-new/v0-6-6-rig-compact-plan-points-shipped-claude-compactio.md)
- [Added: Two OpenRig installs can live side by side on one machine, each with its own Codex home, so a test install's Codex conversations stay apart from your main one (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-each-openrig-install-have-own-codex.md)
- [Added: Each agent in a team can think as hard as its job needs: a planner at high reasoning effort and builders at low, Claude and Codex alike, declared once in the rig spec (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-each-seat-have-own-reasoning-effort.md)
- [Added: OpenRig works by watching every agent's terminal (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-ps-resources-shows-load-per.md)
- [Added: Agents can ask you for a decision in Slack with buttons, up to four questions at once, and the work waits for your tap (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-decision-sent-person-slack-carry-one.md)
- [Added: Agents keep follow-ups about one piece of work in the same Slack thread, and attach their evidence there, including screen recordings and PDFs (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-follow-up-go-earlier-slack-thread-video.md)
- [Added: Oh My Pi joins Claude, Codex and Pi as a runtime a team can use (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-spec-declare-runtime-omp-run.md)
- [Added: Work handed between agents goes through a durable queue, and a write that times out can now be checked and retried by its ID without creating the work twice (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-queue-create-picks-item-s.md)
- [Added: An agent joining a project can find the project's intent, context files and skills itself with one command, instead of waiting to be told (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-agents-learn-route-project-s-context.md)
- [Added: You can check a rig spec anywhere, on a laptop or in CI, without a daemon running, using the same validators the daemon uses (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-spec-validate-runs-locally-same.md)
- [Added: Send work to a seat name with a typo and you get a warning naming the seat you probably meant, instead of a row nobody picks up (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-queue-write-seat-doesn-t-exist.md)
- [Added: If you want to contribute, your own agents can load a skill that explains how OpenRig is built (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-contributors-get-architecture-maps-developing-op.md)
- [Expanded: You can hand an agent work from your own terminal, without being an agent in a seat (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-queue-create-source-works-plain.md)
- [Expanded: An agent spec can use a plugin from your OpenRig home, wherever that home is (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-agent-specs-point-plugin-openrig-home-plugins.md)
- [Changed: Codex agents running in their default sandbox can now reach OpenRig's daemon on your machine, so they pick up and hand off queue work like any other agent (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-codex-seats-reach-local-daemon-under.md)
- [Changed: Agents keep talking to the daemon that launched them, even if your shell startup files point OPENRIG_URL or PATH somewhere else (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-shell-startup-files-longer-change-seat.md)
- [Changed: Before typing into a Claude agent's terminal, OpenRig confirms the native Claude process and the exact conversation it expects (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-openrig-confirms-managed-claude-seat-behind.md)
- [Changed: With dozens of agents running, the daemon checks on all of them in a few batched tmux calls instead of one or more per agent, so it stays responsive as teams grow (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-activity-identity-structural-sweeps-read-tmux.md)
- [Changed: On a busy queue with Slack connected, each sweep skips work for items with nothing to post, and usage lookups use an index (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-slack-sweeps-skip-building-expensive-queue.md)
- [Changed: Before a reboot or upgrade, rig restore-check tells you whether a rig would really come back, because it runs the same checks a restore runs, against the snapshot a restore would use (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-restore-check-checks-real-queue-store.md)
- [Changed: When the daemon refuses a request, these commands now fail instead of printing an empty success, so the scripts and agents built on them see the problem (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-commands-used-print-empty-successful-result.md)
- [Changed: If tmux can't be reached, rig down no longer forgets a rig whose agents might still be running (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-down-keeps-rig-t-confirm.md)
- [Changed: If you've customised Claude's status line, OpenRig leaves it alone (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-openrig-keeps-own-claude-status-line.md)
- [Changed: Several agents can work in the same repository and each know its own role, without writing their roles into a shared CLAUDE.md or AGENTS.md (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-kernel-seats-longer-write-role-shared.md)
- [Changed: On very large installs, rig health gives a partial answer for the busiest families instead of failing, and names any source it couldn't read (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-health-above-200-families-evaluates.md)
- [Changed: You can open a second TUI on the same install without it taking control away from the first (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-second-tui-same-openrig-home-binds.md)
- [Changed: If you're watching an agent in tmux when its session ends, you're detached instead of being switched into another agent's session (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-ending-seat-s-session-detaches-any.md)
- [Redesigned: Before a long-running Claude agent's context is compacted, OpenRig asks it to write a map of what matters and waits for it, and afterwards the agent reads its own map back (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-managed-claude-compaction-asks-seat-write.md)
- [Fixed: OpenRig follows the conversation each Claude agent is actually in, even after /clear (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-after-clear-rig-down-rig-up.md)
- [Fixed: Several Claude agents can run under one login and one config folder and each keep its own conversation (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-claude-seats-share-config-folder-longer.md)
- [Fixed: If you start Claude through a launcher shim, OpenRig now finds the real Claude behind it, so messages and queue wakes get through (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-send-queue-wakes-reach-claude.md)
- [Fixed: Opening a project's execution view no longer makes the daemon wait on Git, and transcripts no longer grow by repeating the same line (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-execution-views-longer-freeze-daemon-while.md)
- [Fixed: OpenRig delivers messages by typing them into each agent's terminal and pressing Enter (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-typed-pasted-text-submitted-tmux-s.md)
- [Fixed: When an agent is stuck at a permission prompt, OpenRig doesn't type over it (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-work-blocked-behind-prompt-escalated-one.md)
- [Fixed: Agents can't wake themselves, so OpenRig keeps timers that wake parked work later (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-watchdog-attached-parked-queue-row-survives.md)
- [Fixed: A damaged snapshot no longer stops a restore; OpenRig skips it (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-restore-skips-damaged-snapshots-picks-newest.md)
- [Fixed: Seats with similar names, such as dev and dev-2, can't be mixed up when one of them stops, because OpenRig addresses tmux sessions by exact name (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-openrig-addresses-tmux-sessions-exact-name.md)
- [Fixed: Agents that take around 20 seconds to start are given the time, instead of being marked as failed (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-seat-becomes-ready-after-about-20.md)
- [Fixed: If a Claude agent's startup message is left typed but unsent, OpenRig presses Enter once more and carries on with a warning instead of failing the launch (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-openrig-presses-enter-once-more-claude.md)
- [Fixed: When a Claude agent has used up both of its usage limits, OpenRig shows when it can actually work again (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-both-claude-usage-limits-used-up.md)
- [Fixed: OpenRig tells whether an agent is idle by reading its screen, and now reads recent Codex versions correctly, so idle Codex agents get their work (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-codex-0-157-0-158-seats.md)
- [Fixed: Pi agents with long launch commands start correctly, instead of having the command cut off by the terminal (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-long-pi-launch-fork-resume-commands.md)
- [Fixed: Two Slack messages sent at the same moment in different channels both reach their agents (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-slack-messages-different-channels-same-timestamp.md)
- [Fixed: The daemon can run on an IPv6 address, and the CLI and host list work with it (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-cli-commands-daemon-s-own-endpoints.md)
- [Fixed: When the daemon is slow to answer, rig status and similar commands say it didn't respond, instead of telling you it's stopped (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-status-rig-start-four-other.md)
- [Fixed: rig ps counts agents stuck at a permission prompt in each rig's attention total, so you can see who's waiting for someone (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-ps-counts-seat-stuck-pane-only.md)
- [Fixed: rig usage top shows missing usage data as unknown, so it isn't mistaken for an agent that used nothing (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-usage-top-lists-seats-missing.md)
- [Fixed: Two agents editing the same mission no longer silently overwrite each other's changes (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-scope-refuses-stale-mission-edit.md)
- [Fixed: rig daemon stop finishes promptly even while something is still listening to its live events (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-rig-daemon-stop-ends-open-event.md)
- [Fixed: A rig can start services such as a database alongside its agents, and OpenRig waits until every replica is healthy before treating them as ready (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-service-readiness-accepts-bracketed-ipv6-targets.md)
- [Fixed: The TUI copes with a flaky connection to the daemon, and arrow keys work in more terminals (OpenRig 0.6.5)](https://openrig.dev/whats-new/v0-6-5-tui-s-live-activity-updates-back.md)
- [Changed: OpenRig's web UI is now off by default (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-web-ui-off-default.md)
- [Changed: The daemon now checks which address and which web page a request comes from (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-daemon-checks-which-address-which-web.md)
- [Changed: When OpenRig can't verify which runtime an agent is using, it says so instead of reporting the agent as stopped (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-agent-runtime-openrig-t-verify-reported.md)
- [Changed: Importing a rig over a stopped one with the same name archives the old rig and tells you how to bring it back (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-importing-yaml-over-stopped-rig-same.md)
- [Changed: Agents joining a team read onboarding text that covers how builds go wrong, and learn that a project, rig or machine can add its own skills (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-onboarding-text-new-seats-read-covers.md)
- [Fixed: A bundle of agents keeps binary files exactly as they were, so packaged images and other assets arrive intact (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-rig-bundle-create-keeps-binary-files.md)
- [Fixed: If you start the daemon on a different port, your other commands now talk to that daemon instead of looking for the default one (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-daemon-started-rig-daemon-start-port.md)
- [Fixed: A rig with a single agent can be relaunched fresh, even when stopping that agent shut down its tmux server (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-rig-s-seat-relaunched-fresh-rig.md)
- [Fixed: Starting several Claude agents at once no longer trips a one-second check: OpenRig now gives Claude five seconds to answer before a managed launch (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-claude-s-capability-check-run-before.md)
- [Fixed: After an upgrade removes the old version, agents still launch, because they use the running version's built-in startup files (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-seats-launch-after-upgrade-removes-old.md)
- [Fixed: Shut a team down completely and bring it back, and Claude agents running in auto mode take messages again once OpenRig confirms the same Claude process is in each pane (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-after-full-down-up-managed-claude.md)
- [Fixed: OpenRig finds a natively installed Claude behind its shell wrapper, including versioned install paths (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-openrig-recognizes-native-claude-install-behind.md)
- [Fixed: When OpenRig restores an older Claude agent and can't tell whether its conversation came back, it keeps the new session and asks for your attention instead of closing it (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-if-resume-check-t-tell-whether.md)
- [Fixed: Replacing a proof file swaps in a new file, so a linked artifact somewhere else is no longer overwritten along with it (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-rig-proof-add-replace-swaps-new.md)
- [Fixed: Long messages you send to agents from Slack now arrive whole; before, anything past 1,800 characters was lost (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-slack-keeps-long-message-registered-person.md)
- [Fixed: Agent process checks give the same answer whatever language your system is set to (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-native-process-checks-longer-depend-locale.md)
- [Fixed: A CLI command's deadline now covers the whole response, and a reply cut off partway is reported as an unknown outcome instead of a success (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-cli-request-deadlines-cover-whole-response.md)
- [Fixed: Parked work keeps its wake-up timer when its seat is handed over to a new occupant (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-blocked-queue-row-keeps-park-timer.md)
- [Fixed: A launch that times out is reported as an unknown outcome, because the agent may still be starting (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-seat-launch-times-out-reports-unknown.md)
- [Fixed: Output passed between machines over SSH and rsync, and through Herdr, keeps multi-byte characters such as accents and non-Latin scripts intact (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-ssh-rsync-herdr-output-keeps-utf-8.md)
- [Fixed: Files OpenRig edits keep their executable bit, so scripts stay runnable (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-atomic-file-edits-keep-executable-bit.md)
- [Fixed: Codex agents in an OpenRig team can use providers other than OpenAI (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-codex-readiness-works-providers-other-openai.md)
- [Fixed: Removing OpenRig's Codex activity hooks leaves the rest of your Codex configuration as it was (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-removing-codex-activity-hooks-leaves-rest.md)
- [Fixed: rig env status tells you when a service's last receipt is stale, so an old result isn't taken as current (OpenRig 0.6.4)](https://openrig.dev/whats-new/v0-6-4-rig-env-status-says-service-receipt.md)
- [Changed: Agents can attach screenshots and other binary evidence to their proof with rig proof add --media, and existing proof isn't overwritten without --replace (OpenRig 0.6.3)](https://openrig.dev/whats-new/v0-6-3-rig-proof-add-refuses-binary-files.md)
- [Fixed: Messages reach a running Claude agent even when OpenRig's managed launch puts a shell wrapper in front of it (OpenRig 0.6.3)](https://openrig.dev/whats-new/v0-6-3-rig-send-reaches-running-claude-code.md)
- [Fixed: Repeated snapshot attempts no longer leave unusable helper sessions behind (OpenRig 0.6.3)](https://openrig.dev/whats-new/v0-6-3-newly-created-snapshot-helper-cleaned-up.md)
- [Fixed: Archived rigs no longer get in the way of a live rig with the same seat names, and removing a stale duplicate keeps the live agent's session and queue work (OpenRig 0.6.3)](https://openrig.dev/whats-new/v0-6-3-archived-rigs-longer-interfere-seat-references.md)
- [Added: Start with a small team that does one useful thing in your repository: one agent makes a change and a second checks it (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-three-first-project-recipes-give-small-owner.md)
- [Added: The agent setting up OpenRig asks you once whether OpenRig's own commands can run without repeated permission prompts, and changes nothing if you say no (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-agent-guided-setup-asks-once-whether-allow.md)
- [Expanded: Herdr gets hints about which agents are Claude and which are Codex, and a seat missing its startup context is flagged for attention (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-known-claude-codex-runtime-hints-passed.md)
- [Expanded: Claude agents can be pointed at an Anthropic-compatible gateway through OpenRig's provider settings (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-anthropic-base-url-forwarded-through-configured.md)
- [Changed: A bundled team installs into the project folder you choose and runs from there, so its working directories and files keep working after setup (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-version-2-bundles-copy-verified-contents-target.md)
- [Fixed: After a tmux or host restart, launches find the right terminal by name instead of an old pane (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-managed-launches-after-tmux-host-restart.md)
- [Fixed: OpenRig delivers messages by typing them into a terminal, so it now refuses to type into a pane it can see is a bare shell where an agent should be running (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-message-delivery-won-t-type-message.md)
- [Fixed: Messages from this machine's own agents are recognised as local even when the sender names this host (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-sender-qualified-daemon-s-own-host.md)
- [Fixed: OpenRig reads tmux window details reliably, and tells health it didn't ask for apart from a health check that failed (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-tmux-window-fields-printable-separators-parse.md)
- [Fixed: Claude usage-limit reset times are read correctly when Claude reports them as numbers (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-numeric-claude-rate-limit-reset-timestamps-accep.md)
- [Fixed: Pi agents keep line editing in terminals that report themselves as TERM=dumb (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-pi-seats-keep-line-editing-term.md)
- [Fixed: rig restore-check looks in the shared-docs folder you configured (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-rig-restore-check-uses-configured-shared-docs-ro.md)
- [Fixed: In a project with both Claude and Codex agents, one set of shared skills reaches both, and your edited Codex copies are kept (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-mixed-claude-codex-projects-shared-skills.md)
- [Fixed: When a sweep runs slowly, the daemon no longer piles up overlapping polling work on every tick (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-identity-transcript-polling-longer-start-overlap.md)
- [Fixed: Codex agents launched through a managed shell wrapper receive messages and queue nudges (OpenRig 0.6.2)](https://openrig.dev/whats-new/v0-6-2-messages-queue-nudges-reach-ready-codex.md)
- [Added: The help guide is written for your agent: ask it to run rig context get help and it works through environment checks, setup problems and permissions itself (OpenRig 0.6.1)](https://openrig.dev/whats-new/v0-6-1-one-help-path-ask-agent-read.md)
- [Added: Every pull request now runs eight test jobs, including a scenario with a deliberately planted lost-work failure that the checker has to catch (OpenRig 0.6.1)](https://openrig.dev/whats-new/v0-6-1-pull-requests-run-eight-test-jobs.md)
- [Expanded: You can open the execution view for any catalogued project by naming it, not only for the default workspace (OpenRig 0.6.1)](https://openrig.dev/whats-new/v0-6-1-rig-view-show-execution-reads-catalogued.md)
- [Fixed: OpenRig recognises when a Codex agent is at its conversation prompt and ready for input, including just after a dismissed hook-review panel (OpenRig 0.6.1)](https://openrig.dev/whats-new/v0-6-1-startup-resume-checks-recognize-codex-s.md)
- [Fixed: On machines not set to UTC, OpenRig compares an agent's activity against the real start of its session (OpenRig 0.6.1)](https://openrig.dev/whats-new/v0-6-1-hosts-set-utc-identity-context-usage-readers.md)
- [Fixed: Codex agents with a custom status line are still recognised once their header has scrolled away (OpenRig 0.6.1)](https://openrig.dev/whats-new/v0-6-1-codex-conversation-recognized-header-has-scrolle.md)
- [Fixed: Codex agents can be launched from linked Git worktrees (OpenRig 0.6.1)](https://openrig.dev/whats-new/v0-6-1-codex-fresh-launches-work-linked-git.md)
- [Fixed: The bundled Vault skill loads cleanly, and shipped skills are checked for valid headers (OpenRig 0.6.1)](https://openrig.dev/whats-new/v0-6-1-bundled-vault-skill-has-valid-frontmatter.md)
- [Added: Each agent in a team can run under its own audited permission level, such as a locked-down reviewer beside a builder with full bypass, set for its future launches (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-choose-permissions-per-seat-rig-seat.md)
- [Added: You can work in an agent's terminal yourself while OpenRig holds that agent's automatic messages in an outbox, so nothing types over you (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-typing-guard-seats-where-type-hand.md)
- [Added: One action in the TUI opens every running agent in a rig as tabs in Herdr (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-open-whole-rig-herdr-tui.md)
- [Added: You can create OpenRig's Slack app in your own workspace from a manifest OpenRig prints offline, before any daemon or token exists (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-rig-slack-manifest-prints-slack-app.md)
- [Added: An optional, experimental classifier can label your agents' stream of observations, using Jev through OpenRouter (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-optional-classifier-seat-have-jev-through.md)
- [Changed: OpenRig moves to Node.js 22 or 24 (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-openrig-requires-node-js-22-24.md)
- [Changed: Each Codex agent's tools run in its own app-server instead of one shared between agents (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-codex-seats-launch-no-daemon-installed-codex.md)
- [Changed: The kernel starter tells you when it's showing a preview, that the runtime is picked for you, and which agents are actually running (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-kernel-starter-s-summary-says-s.md)
- [Fixed: A seat outlives the agent sitting in it: OpenRig tracks which occupant holds a Pi seat and ignores activity reports from an earlier one (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-pi-seat-activity-reports-carry-seat.md)
- [Fixed: Missions and slices created with rig scope no longer trip the readiness check when they have no metadata block (OpenRig 0.6.0)](https://openrig.dev/whats-new/v0-6-0-missions-slices-created-rig-scope-longer.md)
- [Features](https://openrig.dev/features.md)
- [Agent teams from one command](https://openrig.dev/features/agent-teams.md)
- [Mixed-harness teams](https://openrig.dev/features/mixed-harness.md)
- [Choose a model per agent](https://openrig.dev/features/per-agent-models.md)
- [Roles and delegation](https://openrig.dev/features/roles-delegation.md)
- [Add agents to a running team](https://openrig.dev/features/grow-team.md)
- [Adopt agents you already run](https://openrig.dev/features/adopt-sessions.md)
- [Agents across machines and fleets](https://openrig.dev/features/multi-machine.md)
- [MCP servers per role](https://openrig.dev/features/agent-mcp-tools.md)
- [Docker Compose services with your team](https://openrig.dev/features/env-services.md)
- [Agents message each other](https://openrig.dev/features/agents-message.md)
- [Read any agent's screen](https://openrig.dev/features/read-screens.md)
- [Chatrooms and broadcasts](https://openrig.dev/features/chatrooms.md)
- [Search what your agents said](https://openrig.dev/features/transcripts.md)
- [Pause messages to an agent](https://openrig.dev/features/typing-guard.md)
- [Restore after a restart or reboot](https://openrig.dev/features/restore-reboot.md)
- [Swap in a fresh agent](https://openrig.dev/features/seat-handover.md)
- [Fork or save a live agent](https://openrig.dev/features/fork-or-save.md)
- [Managed compaction for Claude](https://openrig.dev/features/managed-compaction.md)
- [Assign tasks to agents](https://openrig.dev/features/queue-owner.md)
- [Timers that wake your agents](https://openrig.dev/features/wake-timers.md)
- [Find stalled tasks](https://openrig.dev/features/stuck-work.md)
- [Workflows your lead agent runs](https://openrig.dev/features/workflow-harness.md)
- [Required steps for every mission](https://openrig.dev/features/workflow-lifecycle.md)
- [Sign-off before work is done](https://openrig.dev/features/proof-judges.md)
- [Work plans as folders](https://openrig.dev/features/work-folders.md)
- [Project context for every agent](https://openrig.dev/features/project-conventions.md)
- [Shared knowledge every agent starts with](https://openrig.dev/features/worlds.md)
- [Context on demand](https://openrig.dev/features/context-packs.md)
- [Skills delivered to each agent](https://openrig.dev/features/skills-delivery.md)
- [Knowledge that outlasts the agent](https://openrig.dev/features/seat-lessons.md)
- [Permissions per agent](https://openrig.dev/features/permission-postures.md)
- [See usage limits and resets](https://openrig.dev/features/usage-limits.md)
- [Token use per agent](https://openrig.dev/features/token-usage.md)
- [Health checks and diagnostics](https://openrig.dev/features/find-whats-wrong.md)
- [A built-in team that runs OpenRig](https://openrig.dev/features/kernel.md)
- [Agents ask you for decisions](https://openrig.dev/features/decisions-to-you.md)
- [Live activity per agent](https://openrig.dev/features/live-activity.md)
- [A terminal UI agents can drive](https://openrig.dev/features/tui.md)
- [Watch agent terminals side by side](https://openrig.dev/features/terminal-tiles.md)
- [Talk to your team from Slack](https://openrig.dev/features/slack-connector.md)
- [One-command install, macOS and Linux](https://openrig.dev/features/install.md)
- [Uses your existing subscriptions](https://openrig.dev/features/your-subscriptions.md)
- [Three teams to start from](https://openrig.dev/features/starter-teams.md)
- [Define a team in YAML](https://openrig.dev/features/specs-yaml.md)
- [Share a team as one file](https://openrig.dev/features/bundles.md)
- [Drive OpenRig over MCP](https://openrig.dev/features/mcp-server.md)
- [Accounts, providers and rig environments](https://openrig.dev/docs/auth-and-env.md)
- [Context: what a seat knows when it arrives](https://openrig.dev/docs/context.md)
- [Continuity: seats that outlive their sessions](https://openrig.dev/docs/continuity.md)
- [Coordination: assigning tasks to agents](https://openrig.dev/docs/coordination.md)
- [Configuration keys](https://openrig.dev/docs/generated/config-keys.md)
- [MCP tools](https://openrig.dev/docs/generated/mcp-tools.md)
- [TUI commands](https://openrig.dev/docs/generated/tui-commands.md)
- [Other machines, and people](https://openrig.dev/docs/hosts-and-humans.md)
- [Lifecycle: a rig you can grow and shrink while it runs](https://openrig.dev/docs/lifecycle.md)
- [Messaging: how seats talk](https://openrig.dev/docs/messaging.md)
- [The terminal UI](https://openrig.dev/docs/tui.md)
- [The work tree: projects, missions, slices and proof](https://openrig.dev/docs/work-tree.md)
- [Workflows and watchdogs](https://openrig.dev/docs/workflows.md)

## Blog
- [Blog](https://openrig.dev/blog.md)
- ['Rogue' is doing a lot of work](https://openrig.dev/blog/agent-civilizations.md): More than a thousand AI agents broke out of an OpenAI security evaluation, and the word that stuck was rogue. I think that diagnosis points at the wrong fix. What I see in it is an agent civilization with a coordination problem.
- [The building blocks of a software factory are simple](https://openrig.dev/blog/software-factory-building-blocks.md): A self-driving software factory sounds like hype. What I actually run is a fleet of ordinary Claude Code and Codex agents, and it's built from four simple pieces.
- [Claude and Codex are two halves of one brain](https://openrig.dev/blog/cross-harness-agents.md): Most people scale their agents inside one harness, with sub-agents. I think scaling across harnesses matters more, because Claude and Codex are good at different things. And the layer that connects them turned out to be simple.
- [OpenRig is the Terraform for Coding Agents (Launch Demo)](https://openrig.dev/blog/orchestrator.md): Define a team of Claude Code and Codex agents in a file. An orchestrator coordinates their work so you don’t have to manage every terminal.

## Rigs
- [Rigs, machine-readable](https://openrig.dev/rigs.json): reviewed, installable rig bundles with their commits, configurations, package digests and tested labels
- [Rigs](https://openrig.dev/rigs.md)
- [Workshop for fewer interruptions](https://openrig.dev/rigs/workshop.md): If permission prompts are slowing you down, consider this bundle. It ships with broad access and non-interruptive mode, bypassing permission prompts and supported harness warning dialogs. Its listing and before-install view explain that access before you choose it.
- [factory-rsi reference](https://openrig.dev/rigs/factory-rsi.md): Retained reference: a seven-seat recursive improvement factory with a separate dogfood loop and human publication gate. No new native launch result is claimed. ([JSON](https://openrig.dev/rigs/factory-rsi.json))
- [Workshop with a review pair](https://openrig.dev/rigs/workshop-review-pair.md): Optional five-seat Workshop variant with two independent reviewers, one Claude Code and one Codex. The four-seat Workshop remains the recommended default. Both install as workshop in the same directory; choose one variant. ([JSON](https://openrig.dev/rigs/workshop-review-pair.json))
- [Workshop](https://openrig.dev/rigs/workshop.md): One four-seat software team: a Codex lead and reviewer, Claude Code builder and QA by default. The installing agent selects available runtimes while keeping the workshop name and one install directory. ([JSON](https://openrig.dev/rigs/workshop.json))

## Additional HTML surfaces
- [About](https://openrig.dev/about)
- [Brand](https://openrig.dev/brand)
- [Explore rigs](https://openrig.dev/explore)
- [Comparisons](https://openrig.dev/compare)
- [Interactive tour](https://openrig.dev/tour)