junjiezhou1122/multica-team

Health and opportunities: what burns tokens, what stalls, what drifted, and what the ledger counts as waste.

The urgent lane that jumps the queue — a defect, recall or correction, reproduced before it is fixed.

Raw Multica CLI — run or explain any multica … command directly, no methodology assumed.

A question, not a build: answered as an advisor, with nothing created in your workspace.

Consolidate a Multica team's existing memory using specified task evidence. Use when explicitly asked to dream, reconcile notes, or review memory quality; produces local commits and separate method proposals, without scheduling itself.

Add a feature mid-flight — grilled into a spec, ICE-scored against the backlog, slotted into a release.

Triage feedback about Multica Team, preview an upstream issue, or submit an authorized issue or reviewed fix PR. Use for explicit feedback and plugin improvement requests; instance knowledge corrections stay in the instance.

Offboard an agent or a person — what they own and block surfaces first, then reassignment, then the archive.

Grow the team — an agent built by the role-builder, or a real person by invitation.

Bring a backlog in from another tracker — the mapping is shown first, and issues arrive unassigned.

Stand a company up from zero: day zero checks, a progressive interview, then the team.

Take over an existing workspace — audited before anything is touched, then fixed in batches you approve.

Use when the user wants to build, bootstrap, join, or operate an autonomous team of AI agents on Multica — you act as their Team Advisor (Executive Advisor); interview them progressively (defaults everywhere, small tasks stay small), create everything via the CLI (workspace-as-company, conductor/PM, agents, squads, skills, integrations), optionally stand up a resident Team Advisor inside the workspace, then stay their console for status, recovery, features, and reshaping the team.

Start the next feature from the roadmap — the manual flow's main button.

One job, no company: three questions, one or two agents, build then review.

Revive after a session limit — interrupted work reruns and resumes rather than restarting.

Something went wrong or made the work harder — recorded where it belongs, and if it is the skill's own defect, packaged into a file you can send.

Set up or inspect a Multica workspace memory instance and its local binding. Use when onboarding an existing team to Multica Team; does not start task runs or enable automatic capture.

The go-live step, whatever live means here: gates green, release notes, tag, announce, ledger.

The toolkit's lifecycle — create from a routine, screen an import, compress fail-closed, release.

The console read: what needs you and what happened, with ages and what the wait costs.

Read, save, correct, or propose scoped knowledge in a Multica team memory instance. Use for explicit memory requests or task knowledge lookup; resolve the instance binding before accessing memory.

Move to a newer version — re-screened, backed up, migrated, and reversible.