ship-dashboards — Claude Code command: dashboard launch plan
ship-dashboards builds a dashboard launch plan covering visualization, publishing, user enablement, and monitoring. Use it to coordinate the rollout and define what to track afterward.
Apache-2.0 ★ 450 updated 2026-04-03
.claude/commands/ship-dashboards.mdWhat ship-dashboards does
The command moves through requirements review, visualization design, publishing, rollout, and monitoring. Its deliverables include a dashboard spec with a wireframe and metrics dictionary, an enablement kit, an enhancement tracker, and a plan entry saved under .claude/plans.
The JSON plan records stages, task dependencies, error handling, and success measures. The workflow can call on bi-publisher, the visualization-patterns and quality-gates skills, plus tools for BI documentation and embedded-dashboard UI checks. Retries for failed publishes or alerts are capped at two. If adoption, freshness, or alert safeguards fail twice within 48h, the plan calls for escalation and rollback.
What it helps with
- align dashboard metrics, filters, and drill paths
- specify chart layouts and accessibility requirements
- plan publishing, permissions, and data refreshes
- prepare user walkthroughs and feedback channels
- track dashboard usage and data freshness
Example prompts for the agent
- Create a launch plan for a pipeline velocity dashboard in Looker for executives and sales:
/analytics-pipeline-orchestration:ship-dashboards --use_case "pipeline velocity" --tool looker --audience "exec,sales" - Plan a Tableau dashboard rollout for analysts, including role-based access and alert thresholds.
- Draft the publishing, enablement, and monitoring plan for our dashboard.
How to install ship-dashboards
Manually: put the files in place
.claude/commands/ship-dashboards.mdCall it with /ship-dashboards. For all projects use ~/.claude/commands/
Manually: put the files in place
~/.codex/prompts/ship-dashboards.mdCall it with /prompts:ship-dashboards. Codex marks this format deprecated in favour of skills
Manually: put the files in place
.github/prompts/ship-dashboards.prompt.mdManually: put the files in place
.gemini/commands/ship-dashboards.tomlGemini stores commands as TOML: the prompt text goes into the prompt field
Questions about ship-dashboards
How do I install ship-dashboards in Claude Code?
Save the command as .claude/commands/ship-dashboards.md.
What does ship-dashboards do?
It creates a dashboard launch plan, from metric and visualization requirements through publishing, user enablement, and monitoring. It also produces a spec and a plan record in .claude/plans.
What happens if monitoring safeguards are breached again?
If adoption, freshness, or alert conditions fail twice within 48h, the plan calls for escalation to Analytics and GTM leadership and a rollback.
Source text
ship-dashboards.md · Apache-2.0
--- name: ship-dashboards description: Creates a dashboard launch plan with visualization specs, enablement steps, and monitoring. usage: /analytics-pipeline-orchestration:ship-dashboards --use_case "pipeline velocity" --tool looker --audience "exec,sales" --- # Command: ship-dashboards ## Inputs - **use_case** – dashboard or report purpose. - **tool** – BI platform (Looker, Tableau, Mode, Sigma, PowerBI). - **audience** – comma-separated audiences. - **access** – optional role-based access requirements. - **alerts** – optional alert thresholds or anomaly rules. ### GTM Agents Pattern & Plan Checklist > Based on GTM Agents orchestrator blueprint @puerto/plugins/orchestrator/README.md#112-325. - **Pattern selection**: Dashboard launches typically run **pipeline** (requirements → visualization → publishing → enablement → monitoring). If publishing + enablement can proceed in parallel, log a **diamond** segment and merge gate. - **Plan schema**: Save `.claude/plans/plan-<timestamp>.json` with objective, visualization stages, task IDs, parallel groups, dependency graph (models, permissions), error handling, and success metrics (adoption %, freshness, alert SLA). - **Tool hooks**: Reference `docs/gtm-essentials.md` stack—Serena for repo/LookML diffs, Context7 for BI platform docs, Sequential Thinking for rollout retros, Playwright for UI QA of embedded dashboards. - **Guardrails**: Default retry limit = 2 for failed publishes/alerts; escalation path = BI Publisher → Analytics Modeling Lead → RevOps/Exec sponsor. - **Review**: Use `docs/usage-guide.md#orchestration-best-practices-puerto-parity` before execution to confirm agents, dependencies, deliverables. ## Workflow 1. **Requirement Review** – confirm metrics, filters, drill paths, narrative structure. 2. **Visualization Spec** – define layout, chart types, color systems, accessibility notes. 3. **Publishing Steps** – build dashboards, set permissions, schedule refreshes, configure alerts. 4. **Enablement & Rollout** – plan walkthroughs, documentation, office hours, feedback channels. 5. **Monitoring** – track usage analytics, data freshness, dashboard performance, enhancement backlog. ## Outputs - Dashboard spec (wireframe, chart list, metrics dictionary). - Enablement kit (loom/video, doc, FAQ, adoption plan). - Monitoring + enhancement tracker. - Plan JSON entry stored/updated in `.claude/plans` for audit trail. ## Agent/Skill Invocations - `bi-publisher` – leads visualization + enablement. - `visualization-patterns` skill – enforces design best practices. - `quality-gates` skill – ensures data + refresh checks. ## GTM Agents Safeguards - **Fallback agents**: document substitutes (e.g., Analytics Data Strategist covering BI Publisher) when specialists unavailable. - **Escalation triggers**: if adoption, freshness, or alert guardrails breach twice in 48h, escalate to Analytics + GTM leadership and trigger rollback plan. - **Plan maintenance**: update plan JSON/change log whenever visual specs, access rules, or monitoring hooks change to maintain GTM Agents-grade auditability. ---
© gtmagents · Apache-2.0 · gtmagents/gtm-agents