# JoyStream

## Docs

- [Addressing](https://docs.joystream.ai/cli/addressing.md): How to refer to agents, workspaces, runs, and catalog entries — handles, UUIDs, fully qualified names, and sequence addresses.
- [jstm admin](https://docs.joystream.ai/cli/admin.md): User administration for self-hosted operators.
- [jstm agent build](https://docs.joystream.ai/cli/agent-build.md): Compile a committed spec into an immutable, versioned agent package.
- [jstm agent conform](https://docs.joystream.ai/cli/agent-conform.md): Check a built agent package against the agent spec — flags, lanes, ignores, JSON schema, and exit codes.
- [Inspect an agent](https://docs.joystream.ai/cli/agent-inspect.md): list, show, view, status, spine, spec, readiness, and readiness-fix.
- [Manage agents](https://docs.joystream.ai/cli/agent-manage.md): search, install, visibility, copy, move, duplicate, and delete.
- [jstm agent plan](https://docs.joystream.ai/cli/agent-plan.md): Draft and edit agent specs in an interactive chat. Drafts stay drafts until plan done commits them.
- [jstm agent run](https://docs.joystream.ai/cli/agent-run.md): Run an agent, stream its progress, and read the result.
- [jstm agent schedule](https://docs.joystream.ai/cli/agent-schedule.md): Show, set, and clear an installation's cron schedule. All times are UTC.
- [jstm agent dry-run, pilot, live](https://docs.joystream.ai/cli/agent-stages.md): Stage an agent and promote it through Dry Run, Pilot, and Live.
- [Authentication](https://docs.joystream.ai/cli/auth.md): jstm init, login, logout, and whoami — every sign-in mode and what each stores.
- [Configuration](https://docs.joystream.ai/cli/configuration.md): Where the CLI stores its config and session, which environment variables it reads, and how each value is resolved.
- [jstm connection](https://docs.joystream.ai/cli/connection.md): Connect services to a workspace, test them, and see how a credential resolves.
- [jstm credits](https://docs.joystream.ai/cli/credits.md): Check a workspace's credit balance and recent ledger entries.
- [jstm input-requests](https://docs.joystream.ai/cli/input-requests.md): Answer an agent that has paused waiting for information from a person.
- [jstm mcp](https://docs.joystream.ai/cli/mcp.md): Browse the MCP server catalog, register your own servers, and manage source repositories.
- [jstm org](https://docs.joystream.ai/cli/org.md): List organizations, view members, and invite or remove people.
- [CLI overview](https://docs.joystream.ai/cli/overview.md): How the jstm command line works — global flags, JSON output, streams, and exit codes.
- [jstm profile](https://docs.joystream.ai/cli/profile.md): View or update your profile, and manage your email addresses.
- [jstm registry](https://docs.joystream.ai/cli/registry.md): See the resolved catalog for a workspace, and find out which entry wins when names collide.
- [jstm session](https://docs.joystream.ai/cli/session.md): Pin catalog skills to a builder session so Joyvis builds around them.
- [jstm skills](https://docs.joystream.ai/cli/skills.md): Browse the skills catalog, check whether a skill can run, and manage skill source repositories.
- [jstm ticket](https://docs.joystream.ai/cli/ticket.md): Read a run's work as tickets — the business-readable view of what an agent did.
- [jstm triggers](https://docs.joystream.ai/cli/triggers.md): Bind an agent to an external event — create webhook triggers, get their URL, send a signed test, and inspect delivered events.
- [jstm vault](https://docs.joystream.ai/cli/vault.md): Vaults and credentials — list, create, inspect, audit across the org, and set availability scope.
- [jstm workspace](https://docs.joystream.ai/cli/workspace.md): Create, inspect, and switch workspaces, and manage their members.
- [Agent Engineering](https://docs.joystream.ai/concepts/agent-engineering.md): Generating an agent is getting easy. Trusting one is not. Agent Engineering is the discipline that closes the gap.
- [Agents and recipes](https://docs.joystream.ai/concepts/agents-and-recipes.md): An agent is the governed intelligent realization of a capability spec. The recipe is the approved route it travels.
- [Capabilities](https://docs.joystream.ai/concepts/capabilities.md): The primary object in JoyStream — the durable organizational ability a team owns, operates, and continuously improves.
- [Capability spec](https://docs.joystream.ai/concepts/capability-spec.md): The declarative definition of a capability — what must be true, within what boundaries, and how success will be proven.
- [Evidence, memory, and metrics](https://docs.joystream.ai/concepts/evidence-and-memory.md): What counts as proof that a capability worked, what an agent is allowed to learn, and how outcomes are measured.
- [The lifecycle](https://docs.joystream.ai/concepts/lifecycle.md): Plan, Build, Conform, Dry Run, Pilot, Live — the stages an agent passes through, and the two loops that compound around them.
- [How JoyStream thinks](https://docs.joystream.ai/concepts/overview.md): The belief, the category, and the primary object — the chain of reasoning the whole product rests on.
- [Runs and work](https://docs.joystream.ai/concepts/runs-and-work.md): A run materializes a recipe as an epic of stories, and surfaces them as tickets people can read, own, and resolve.
- [Skills and MCP servers](https://docs.joystream.ai/concepts/skills-and-mcp.md): How an agent reaches the outside world — skills, MCP servers, catalogs, repositories, and scope resolution.
- [Terminology](https://docs.joystream.ai/concepts/terminology.md): The canonical vocabulary, and how each term maps to what you see in the CLI and the product today.
- [Workspaces and organizations](https://docs.joystream.ai/concepts/workspaces-and-organizations.md): The boundary in which capabilities are owned and operated — workspace types, roles, access modes, and agent visibility.
- [Agent inputs](https://docs.joystream.ai/features/agent-inputs.md): How an agent declares what it needs, what happens when something is missing, and how a run resumes once you supply it.
- [Planning an agent in chat](https://docs.joystream.ai/features/agent-plan-chat.md): Draft and edit agent specs in an interactive chat. Your work stays a draft until you explicitly commit it.
- [Reading the flag report](https://docs.joystream.ai/features/conformance/flag-report.md): How to read jstm agent conform output — lanes, severity, flag anatomy, JSON schema, and exit codes.
- [Acknowledging flags](https://docs.joystream.ai/features/conformance/ignores.md): See it, decide, move ahead — acknowledge a conformance flag so it stops blocking without disappearing from the record.
- [Agent package conformance](https://docs.joystream.ai/features/conformance/overview.md): Check that a built agent package provably meets the agent spec — a fast, deterministic gate between Build and Dry Run that needs no API key.
- [Credentials and vaults](https://docs.joystream.ai/features/credentials-and-vaults.md): Where JoyStream stores service credentials, how org admins audit them, how credential scope controls availability, and how tokens reach an MCP server.
- [Credits](https://docs.joystream.ai/features/credits.md): How JoyStream meters consumption — balance, ledger, and what draws credits down.
- [Importing private GitHub repositories](https://docs.joystream.ai/features/private-repos.md): Import private GitHub repos into your skill sources and MCP catalog using your own GitHub credential.
- [Scheduling agents with cron](https://docs.joystream.ai/features/scheduling.md): Run an installed agent automatically on a cron schedule — how schedules fire, dry_run versus live, and what last-fired and next-fire mean.
- [Sharing](https://docs.joystream.ai/features/sharing.md): Send someone a link to an agent, a run, or a runs list — how the invitation works and what accepting grants.
- [Triggers](https://docs.joystream.ai/features/triggers.md): Run an agent from an external event — webhook bindings, HMAC signing, and the event record.
- [Visibility and discovery](https://docs.joystream.ai/features/visibility-and-discovery.md): How a good agent spreads across an organization — visibility levels, the registry, and installations.
- [Add skill and MCP repositories](https://docs.joystream.ai/guides/add-skill-and-mcp-repositories.md): Sync a GitHub repository of skills or MCP servers into your catalog, and control who sees it.
- [Add your own MCP server](https://docs.joystream.ai/guides/add-your-own-mcp-server.md): Register an MCP server you host or subscribe to, attach its key, and confirm its tools are available.
- [Audit and scope org credentials](https://docs.joystream.ai/guides/audit-and-scope-org-credentials.md): Review every credential across the organization and control which workspaces each one is available to, in the UI and the CLI.
- [Build your first agent](https://docs.joystream.ai/guides/build-your-first-agent.md): The full path from an idea to a package that has run end to end — plan, commit, build, conform, dry run.
- [Conform an agent package](https://docs.joystream.ai/guides/conform-an-agent-package.md): Run conformance on a built package, read the flags, fix or acknowledge them, and go green.
- [Connect a service](https://docs.joystream.ai/guides/connect-a-service.md): Give a workspace the credentials its agents need, and confirm an agent will resolve the one you intended.
- [Connect GitHub for private repos](https://docs.joystream.ai/guides/connect-github-private-repos.md): Connect a repo-scoped GitHub token and import a private repository as a skill source or MCP repository.
- [Draft an agent with chat](https://docs.joystream.ai/guides/draft-an-agent.md): Create or edit an agent spec in the plan chat, then commit it.
- [Invite and manage members](https://docs.joystream.ai/guides/invite-and-manage-members.md): Add people to an organization and its workspaces, choose their roles, and remove access.
- [Manage credits](https://docs.joystream.ai/guides/manage-credits.md): Check a workspace's credit balance and find out where credits went.
- [Organize and publish agents](https://docs.joystream.ai/guides/organize-and-publish-agents.md): Move an agent to the team that owns the work, make it discoverable, and let other teams install it.
- [Promote to production](https://docs.joystream.ai/guides/promote-to-production.md): Move an agent from Dry Run through Pilot to Live, and what evidence to check at each gate.
- [Review runs and tickets](https://docs.joystream.ai/guides/review-runs-and-tickets.md): Read what an agent actually did — runs, stories, tickets, and the provenance behind them.
- [Run an agent](https://docs.joystream.ai/guides/run-an-agent.md): Trigger a run, follow its progress, pass input, and unblock it when it needs a person.
- [Schedule an agent](https://docs.joystream.ai/guides/schedule-an-agent.md): Put an installed agent on a recurring cron schedule, verify it, switch it to live, and clear it.
- [Set up your organization](https://docs.joystream.ai/guides/set-up-your-organization.md): Create an organization, add team workspaces, and move between them.
- [Share agents and runs](https://docs.joystream.ai/guides/share-agents-and-runs.md): Send someone a link to an agent, a run, or a workspace's run list — including people who aren't in your workspace yet.
- [Trigger an agent](https://docs.joystream.ai/guides/trigger-an-agent.md): Run an agent from an external event — create a webhook binding, test it, and inspect deliveries.
- [JoyStream](https://docs.joystream.ai/index.md): The operating system for organizational agents — engineer, operate, and improve the capabilities that run your business.
- [Install the CLI](https://docs.joystream.ai/installation.md): Get the jstm command-line client on your machine and sign in.
- [Quickstart](https://docs.joystream.ai/quickstart.md): Draft, build, and dry-run your first agent from the terminal.
- [Common errors](https://docs.joystream.ai/troubleshooting/common-errors.md): What each CLI error means and what to do about it.
- [Conformance failures](https://docs.joystream.ai/troubleshooting/conformance.md): Why a conform run failed, and what to do about each cause.
- [Credential problems](https://docs.joystream.ai/troubleshooting/credentials.md): Missing credentials, GitHub scopes, resolution surprises, and what each error contract means.
- [FAQ](https://docs.joystream.ai/troubleshooting/faq.md): Common questions about JoyStream, agents, credentials, and the CLI.
- [Blocked or failing runs](https://docs.joystream.ai/troubleshooting/runs.md): A run that stopped, stalled, never started, or finished wrong — how to find out which, and what to do.
- [Use cases](https://docs.joystream.ai/use-cases/overview.md): Five organizational capabilities, and what building each one in JoyStream looks like.