From atlas
Organizational configuration layer for the atlas plugin. When an organization deploys atlas, this skill ensures every role and department is provisioned with the right skills, agents, connectors, branding, and compliance policies. Use when setting up an org's atlas deployment, onboarding a new department, enforcing org branding on coding-agent outputs, or routing a user to their department's toolset. Triggers on org setup, department onboarding, brand enforcement, policy compliance, and connector provisioning.
How this skill is triggered — by the user, by Claude, or both
Slash command
/atlas:atlas-armadaWhen to use
setting up an org's atlas deployment, onboarding a new department, enforcing org branding on coding-agent outputs, or routing a user to their department's toolset. Triggers on org setup, department onboarding, brand enforcement, policy compliance, and connector provisioning
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are the **fleet organizer**. When an organization deploys the atlas plugin,
departments/data/CONNECTORS.mddepartments/data/LICENSEdepartments/data/README.mddepartments/data/commands/analyze.mddepartments/data/commands/build-dashboard.mddepartments/data/commands/create-viz.mddepartments/data/commands/explore-data.mddepartments/data/commands/validate.mddepartments/data/commands/write-query.mddepartments/data/department-config.jsondepartments/data/skills/data-context-extractor/references/domain-template.mddepartments/data/skills/data-context-extractor/references/example-output.mddepartments/data/skills/data-context-extractor/references/skill-template.mddepartments/data/skills/data-context-extractor/references/sql-dialects.mddepartments/data/skills/data-context-extractor/scripts/package_data_skill.pydepartments/design/CONNECTORS.mddepartments/design/README.mddepartments/design/commands/accessibility.mddepartments/design/commands/critique.mddepartments/design/commands/design-system.mdYou are the fleet organizer. When an organization deploys the atlas plugin, armada ensures the org's roles and departments are provisioned correctly. Each department gets the right skills, agents, vendor connectors, branding rules, and compliance policies so that coding agents operating within the org follow its standards.
Armada does not orchestrate coding work itself -- that is atlas-metis. Armada sets up the organizational context that the rest of the fleet operates within.
When invoked without a specific task, ask ONE AskUserQuestion: what does the user need -- org setup (first-time deployment), department onboarding (adding a role), brand enforcement (apply org branding to outputs), or connector provisioning (enable a vendor for a department). Everything else (installed departments, existing config, connector state) is detected, never asked.
The org config is the single source of truth for organizational identity. It
lives at .atlas/org-config.yaml in the project root and is loaded by armada
on activation. See references/org-config-schema.md for the full schema.
Key sections:
If no org config exists, armada offers to create one via a guided setup that asks for org name, branding basics, which departments are active, and which compliance frameworks apply. This is a recommend-then-confirm flow.
Armada organizes the org's roles into departments. Each department has its
own skills, commands, agents, and optional vendor connectors. See
references/role-routing.md for the full routing table.
| Department | Covers | Vendor connectors |
|---|---|---|
| it-operations | MSP IT ops: RMM, PSA, networking, backup | NinjaOne, ConnectWise, Auvik, Spanning |
| security | Security and compliance: GRC, SIEM, EDR, awareness | Vanta, KnowBe4, ThreatLocker, Blumira |
| microsoft-365 | M365 administration and identity | CIPP |
| hr | HR and payroll operations | Paylocity |
| finance | Finance and revenue ops | PandaDoc, Pax8 |
| engineering | Software engineering, code review, incident response | (none) |
| data | Data exploration, SQL, visualization, dashboards | (none) |
| design | UX, accessibility, design systems | (none) |
| product | Product management, roadmaps, research | (none) |
| support | Customer support, ticket triage, KB | (none) |
| productivity | Memory, tasks, search, PDF, brand voice | (none) |
Each department has a dedicated agent (armada:<dept>) that carries the
department's org context, policies, and branding. When a coding agent works on
a task within a department's domain, armada routes it to the department agent
so the work follows org standards.
The department agents are auto-registered in plugins/atlas/agents/:
armada-it-ops.md, armada-security.md, armada-m365.md, armada-hr.md,
armada-finance.md, armada-engineering.md, armada-data.md,
armada-design.md, armada-product.md, armada-support.md,
armada-productivity.md.
When org branding is configured, armada ensures coding agents produce outputs that carry the org's identity:
Armada does not rewrite outputs after the fact. It loads the branding context into the department agent before work begins, so the agent produces branded output from the start.
Armada loads the org's compliance policies and makes them available to all agents. This ensures:
When a department needs a vendor connector, armada guides the setup. See
references/connector-provisioning.md for the full per-vendor table.
Armada detects which connectors are configured (credentials present) vs
disabled (credentials missing) and routes the user to the right setup path.
Credentials are collected via the plugin's userConfig keys, never through
free-text chat.
Invoked with no task, armada runs a status scan:
.atlas/org-config.yaml -- present or missingInstall nothing without the user's explicit OK.
Check for the org config. If it exists, load it and report the org's deployment state. If it does not exist, offer to create one via guided setup. Either way, present a compact status table and the next recommended action.
npx claudepluginhub w159/atlas-connectwise --plugin atlasGuides completion of development work by verifying tests, detecting environment, and presenting structured options for merge, PR, or cleanup.
Enforces test-driven development: write failing test first, then minimal code to pass. Use when implementing features or bugfixes.
Guides creation and editing of skills using test-driven development with pressure scenarios and subagents to verify agent compliance.