agency-dashboard
Generate a portfolio-level dashboard across ALL client brands — per-client RAG health scores, campaign activity, budget pacing, aggregate KPIs, team utilization, pending approvals, upcoming deadlines, and an alerts panel — built for agency standups and weekly reviews. Triggers on \"/digital-marketing-pro:agency-dashboard\", \"how are all our clients doing\", \"portfolio health check\", \"budget pacing across accounts\", \"which accounts are at risk\". Enumerates every brand under ~/.claude-marketing/brands/ and pulls data via campaign-tracker.py, execution-tracker.py, and team-manager.py; drill into a single client with /digital-marketing-pro:performance-report or /digital-marketing-pro:client-report.
indranilbanerjee · 已有 AI 审核记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add indranilbanerjee/digital-marketing-pro --skill agency-dashboard技能说明
name: agency-dashboard description: "Generate a portfolio-level dashboard across ALL client brands — per-client RAG health scores, campaign activity, budget pacing, aggregate KPIs, team utilization, pending approvals, upcoming deadlines, and an alerts panel — built for agency standups and weekly reviews. Triggers on "/digital-marketing-pro:agency-dashboard", "how are all our clients doing", "portfolio health check", "budget pacing across accounts", "which accounts are at risk". Enumerates every brand under ~/.claude-marketing/brands/ and pulls data via campaign-tracker.py, execution-tracker.py, and team-manager.py; drill into a single client with /digital-marketing-pro:performance-report or /digital-marketing-pro:client-report." user-invocable: true triggers:
- agency portfolio dashboard
- cross-brand campaign status
- budget pacing all clients
- agency KPI overview
- portfolio health check
- multi-client dashboard
- agency team utilization
- overview of all client accounts
/digital-marketing-pro:agency-dashboard
Purpose
Generate a portfolio-level dashboard aggregating health metrics across ALL client brands. Shows campaign activity, budget pacing, KPI attainment, content pipeline, and team utilization at a glance — giving agency leadership a single view of operational health without opening each account individually. Designed for daily standups, weekly agency reviews, or on-demand health checks.
Input Required
The user must provide (or will be prompted for):
- Dashboard scope: All brands or a specific list of brand slugs to include in the portfolio view
- Time period: Current week, month, or quarter — determines the pacing calculations and comparison windows
- Detail level: Summary (top-line health scores per client) or detailed (campaign-level breakdowns per client with individual campaign metrics)
- Sort/filter preferences: Sort clients by health score, spend, revenue, or alphabetical — and optionally filter to only at-risk (amber/red) accounts
- Team filter (optional): Filter by account lead or team pod if the agency has multiple pods managing different client sets
- Comparison baseline (optional): Compare current period against prior period, same period last year, or plan/target — defaults to prior period
- Alert threshold overrides (optional): Custom thresholds for performance drop alerts or budget pacing tolerance — defaults to 20% performance drop and 10% pacing variance
- Export format (optional): Whether to output as markdown, Google Sheets, or Slack message — defaults to markdown
Process
- Load brand context: Read
~/.claude-marketing/brands/_active-brand.jsonfor the active slug, then load~/.claude-marketing/brands/{slug}/profile.json. Apply brand voice, compliance rules for target markets (skills/context-engine/compliance-rules.md), and industry context. Also check for guidelines at~/.claude-marketing/brands/{slug}/guidelines/_manifest.json— if present, load restrictions. Check for agency SOPs at~/.claude-marketing/sops/. If no brand exists, ask: "Set up a brand first (/digital-marketing-pro:brand-setup)?" — or proceed with defaults. - Enumerate all brands: Scan
~/.claude-marketing/brands/for all configured brand directories (excluding_active-brand.json). For each brand, loadprofile.jsonto get client name, industry, engagement type, contract dates, assigned team members, and KPI targets - Pull campaign data per brand: For each brand in scope, run
python "${CLAUDE_PLUGIN_ROOT}/scripts/campaign-tracker.py" --brand {slug} --action list-campaignsto retrieve active campaigns, statuses, budgets, and objectives (use--action get-campaign --id {id}for a single campaign's detail) - Pull execution status per brand: For each brand, run
python "${CLAUDE_PLUGIN_ROOT}/scripts/execution-tracker.py" --brand {slug} --action get-historyto get logged executions — completed deliverables, launches, and tasks — then derive pending / overdue status in analysis - Check budget pacing per brand: For each brand, compare actual spend-to-date against planned spend for the current period — calculate pacing percentage and project end-of-period spend at current run rate
- Calculate per-client health score: Apply the RAG scoring formula from
skills/context-engine/agency-operations-guide.md:- Green: On track across KPIs, budget on pace (within 10%), no overdue items, content pipeline flowing
- Amber: 1-2 KPIs at risk, minor pacing drift (10-20%), items approaching deadline, or pending approvals aging
- Red: Significant KPI misses, budget overspend (>20%), missed deadlines, stalled campaigns, or MCP disconnections
- Aggregate portfolio KPIs: Sum total active campaigns, total monthly spend, average ROAS across clients, total leads/conversions, total pending deliverables, and overall portfolio health distribution (count and percentage of green/amber/red)
- Check team utilization: Run
python "${CLAUDE_PLUGIN_ROOT}/scripts/team-manager.py" --action check-capacity --brand {slug}to assess current team workload — available capacity per team member, overloaded staff flagged, accounts at risk of under-service, and billable-hours tracking. Optionally add a Claude Code cost line — only if the user supplies the data. The plugin cannot read Claude Code usage itself. If the user pastes the output of their own/usagecommand (a Claude Code CLI slash command they run interactively), aggregate the per-model token/cost figures it reports into a "Claude Code consumption" line so leadership can see AI cost before the monthly invoice. Label models by whatever the user's/usageoutput names — do not assume specific model IDs. Brand-per-directory workspaces (~/work/clients/{slug}) make the figures brand-attributable. - Identify pending approvals: Scan execution logs across all brands for items awaiting client or internal approval — flag anything ol