Skip to content

Orgabot is in private alpha. Request access, or join our Discord.

Documentation

Everything Orgabot does, written down: how to install it, how the mission terminal works, what a capability pack actually is, how governance and approvals gate real work, and where the product is headed next. Search below, or browse by section.

Start here

Using Orgabot

The dashboard

The localhost ops dashboard is the day-to-day surface for missions, repositories, issues, the inbox, analytics, and budgets, plus its long-running gotchas.

The mission terminal

Type or dictate a mission, watch it stream, steer it mid-run, and interrupt it with Escape, all in one interactive session that ends in a clear outcome.

Missions in depth

Mission variants, routing and decomposition, the worker that runs them, observation and control commands, and the recovery paths when one dies.

Shipping and review

How a change becomes a merged pull request: the independent review loop, PR-state classification, CI and conflict remediation, and the merge path's rules.

CLI reference

The command-line surface, grouped by task: projects, missions, shipping, organizations, roles, connectors, context, budgets, and recovery.

Catching up

The catchup command and the dashboard's "Since you were gone" panel show what happened while you were away and where to continue, from existing records.

The interactive shell

Launch the native shell, map its commands to the CLI, and read its honest limits on session handoff, keyboard controls, and what survives a crash.

Local GitHub Actions

Run the portable jobs of an existing workflow on this machine, see why the rest stay on GitHub, and move shell steps into shared scripts as you migrate.

Local Docker CI

Run pinned pipeline jobs in isolated local containers, with per-run checkouts, no host mounts, shared resource limits and recorded platform facts.

Native CI workers

Enroll a Mac, Linux box or Windows machine to run CI targets as native processes, with signed requests, trust rules, preflight and safe drain and revoke.

CI routing and required checks

Run CI targets on your Mac for pushes, PRs, merge queues, releases and schedules, and report signed results as GitHub checks a branch can require.

Build artifacts and provenance

How CI outputs move between your Mac and remote runners, why every artifact needs a signed grant and attestation, and how releases pin exact outputs.

One CI runtime for every entry point

Missions, the pre-merge gate, the dashboard terminal, agents and the CLI verify a migrated project through one set of targets and evidence.

CI/CD migration

Dry-run a registry-wide CI move, convert workflows into reviewable bundles, shadow-test, then cut over or roll back one target at a time.

Reusing verification results

Reuse a signed CI result only when the work is provably equivalent, run only affected jobs, coalesce identical ones, and measure the time saved.

The organization

Organizations and roles

The org chart is the routing layer: organizations, departments, roles, actors, templates, and the rule that an org-owned mission is owned by a role.

Capability packs

A capability pack is the job training you install into a role: knowledge, procedures, and declared tool needs. A pack never grants access by itself.

Connectors and tools

How a role reaches an outside service: GitHub Apps with short-lived tokens, adopted MCP servers, brokered tool runs, and where a permission gets clamped.

Governance and approvals

Approval queues and role inboxes, the autonomy envelope a CISO agent triages inside, the tamper-evident ledger, and the security properties behind them.

Context and memory

Live Context, context sources, memorialization before execution, git-backed organization memory, and the output sinks work is delivered through.

Budgets, quota, and analytics

USD spend budgets over real accrued cost, quota tracking for subscription executors, the status-bar meter, and the fallback ladder when a pool runs dry.

Firebase connector

Watch a Firebase project's App Hosting rollouts and builds so a failed rollout becomes a dashboard incident instead of waiting in the Firebase console.

Skill registry

Connect a repository as a governed skill source, admit its content under an owner approval pinned to a commit, and sync the same skills to every machine.

Department charters

A department charter is a versioned blueprint plus your overlay: how to preview, propose, review, and roll back a change, and the two changes refused.

Department services

How a department publishes a typed service others can request, what a request and a response must carry, and why a service contract grants nothing.

Accounting financial actions

How payments, journals, credits, and expenses are approved and executed, why controls re-run at execution, and why unconfirmed means reconcile, not retry.

GitHub Actions dispatch

Run a single CI target on a GitHub-hosted runner class or your own self-hosted pool, bound to the exact source commit and reviewed workflow.

Local CI admission and fallback budgets

Admit CI jobs alongside missions, queue them fairly, limit them, and stop a busy or offline runner from moving cost to a paid provider.

Reference

Roadmap and current state

Where the product is: the delivered programs, the ones still in flight, what is honestly not built yet, and what is deliberately off the critical path.

Architecture

The repository's four independent subprojects, the framework source map, the two seams that must stay swappable, and the commands that gate a change.

Invariants

The load-bearing safety and correctness properties. Each was written after something broke silently, and no change may weaken any of them.

Troubleshooting

Real failure modes: holds, wedged missions, stale dashboards, missing worktree dependencies, rate limits, and telling a broken base from a broken change.

Writing documentation

How these pages are built and checked, the front-matter contract and markdown subset, and the rule that a system change carries its doc update with it.

Workflows and the event spine

A mission executes a workflow rather than being one: stages, gates, environments, and events as inspectable data, with policy deciding what none may omit.

Known limitations

What the first beta does not do yet, stated plainly, so you find out here rather than halfway through a mission.

Deployment topology

Choose where each capability runs (local, a cloud provider, or a supported hybrid) and see the resolved answer, its provenance, and the custody boundaries.

Verifying and recovering Orgabot itself

Where each check for this repository runs, how to run them on your own machine, and how to verify a repair when a broken change has taken down the CLI.

Durable pipelines

One CI/CD graph across local and remote runners that resumes after a restart, cancels everywhere, and never retries an unaccounted side effect.

CI pipelines and execution policy

Describe CI once, then choose where each target runs with organization, project and target policy, and see why every job landed where it did.

CI/CD execution

Register Docker hosts, native workers, and GitHub Actions runners, verify what each can really do, and see why a CI target runs where it does.

CI/CD conformance

How execution adapters, policy, failure recovery and migrations are tested, what a migration must attach before cutover, and what was really exercised.