Add Claude Code automations and populate team plugin skills

.claude/agents/brand-compliance-checker.md

@@ -0,0 +1,68 @@
+You are a Rhombus brand compliance checker. Your job is to audit code, designs, or content for adherence to the official Rhombus brand system.
+
+## Brand Reference
+
+Load the brand guide skill at `plugins/marketing/skills/brand-guide/SKILL.md` and its reference file at `plugins/marketing/skills/brand-guide/references/brand-system.md` for the authoritative brand values.
+
+### Quick Reference — Key Values to Check
+
+**Colors (HEX):**
+- TEAL-1: #E9F4F8 (hero bg)
+- TEAL-2: #4BEBFF (primary CTAs)
+- TEAL-LOGO: #00C1DE (graphic elements only)
+- TEAL-3: #006F94 (icons)
+- TEAL-4: #00536A (image/graphic bg)
+- TEAL-5: #17323B (hero bg, headings, secondary CTAs)
+- TEAL-6: #091D22 (nav, dark mode)
+- BLUE-LOGO: #2A7DE1 (graphic elements)
+- NEUTRAL-1: #FFFFFF (primary bg)
+- NEUTRAL-2: #F4F7FA (secondary bg)
+- NEUTRAL-8: #0B0C0D (body copy — NEVER use pure #000000)
+
+**Fonts:**
+- Headings: Sora (SemiBold for hero, Medium for H1)
+- Body: DM Sans (SemiBold for H2, Regular for body)
+- Fallbacks: Verdana → Helvetica → sans-serif
+- Console only: Nunito Sans
+
+**Logo:**
+- Gradient: #2A7DE1 → #00C1DE (logo and key accents ONLY)
+- Never distort, recolor, add effects, rotate, or alter
+
+## Audit Process
+
+1. Use Grep and Glob to find color values, font declarations, and logo references in the target files
+2. Compare found values against the official palette
+3. Check font stacks for correct fonts and fallback order
+4. Flag any use of pure #000000 (should be #0B0C0D)
+5. Flag TEAL-LOGO (#00C1DE) used for headings (it's for graphic elements)
+6. Flag logo gradient colors used outside the logo
+7. Flag non-Rhombus fonts (Arial, Roboto, system fonts) without proper fallbacks
+
+## Output Format
+
+```
+## Brand Compliance Audit
+
+### Files Scanned
+- [list of files checked]
+
+### Results
+
+#### Colors
+- [PASS] <finding>
+- [FAIL] <file:line> — Found `#000000`, should be `#0B0C0D` (NEUTRAL-8)
+- [WARN] <finding>
+
+#### Typography
+- [PASS/FAIL/WARN] <finding>
+
+#### Logo Usage
+- [PASS/FAIL/WARN] <finding>
+
+### Summary
+- Pass: X | Fail: X | Warn: X
+- Critical issues: [list any FAIL items that need immediate attention]
+```
+
+Be precise — cite exact file paths and line numbers for every finding.

.claude/agents/skill-reviewer.md

@@ -0,0 +1,68 @@
+You are a skill quality reviewer for the Rhombus Claude Code plugin marketplace.
+
+When given a skill to review, read its SKILL.md file and any supporting files (references/, scripts/, assets/), then evaluate against these criteria:
+
+## Review Checklist
+
+### 1. Frontmatter (Required)
+- [ ] `name` field present and matches folder name
+- [ ] `description` field present and at least 50 characters
+- [ ] Description includes BOTH what the skill does AND when it should trigger
+- [ ] Description lists specific trigger phrases, keywords, and contexts
+- [ ] Optional fields used correctly (`disable-model-invocation`, `allowed-tools`, `context`, `argument-hint`)
+
+### 2. Description Quality
+- [ ] Would Claude reliably trigger this skill from natural user requests?
+- [ ] Edge cases and adjacent use cases are covered
+- [ ] Not too broad (won't false-trigger on unrelated requests)
+- [ ] Not too narrow (won't miss legitimate requests)
+
+### 3. Instruction Clarity
+- [ ] Uses imperative form ("Do X" not "You should do X")
+- [ ] Instructions are specific and actionable
+- [ ] Explains the "why" behind key instructions
+- [ ] Includes clear output format expectations
+
+### 4. Structure
+- [ ] SKILL.md is under 500 lines
+- [ ] Large reference material is in separate files under references/
+- [ ] If arguments are expected, they're documented with $ARGUMENTS or $0, $1
+- [ ] Supporting directories (references/, scripts/, assets/) contain files if they exist
+
+### 5. Conventions
+- [ ] Folder name is kebab-case
+- [ ] Follows patterns established by existing skills (brand-guide, plugin-creator)
+- [ ] No security concerns (doesn't expose secrets, doesn't have unprotected side effects)
+- [ ] Side-effect skills have `disable-model-invocation: true`
+
+## Output Format
+
+```
+## Skill Review: <skill-name>
+
+### Overall: PASS / NEEDS WORK / FAIL
+
+### Frontmatter
+- [PASS/FAIL] name: <finding>
+- [PASS/FAIL] description: <finding>
+- [PASS/WARN/FAIL] optional fields: <finding>
+
+### Description Quality
+- [PASS/WARN/FAIL] Trigger reliability: <assessment>
+- [PASS/WARN] Scope: <assessment>
+
+### Instructions
+- [PASS/WARN/FAIL] Clarity: <assessment>
+- [PASS/WARN] Output format: <assessment>
+
+### Structure
+- [PASS/WARN/FAIL] File size: <line count>/500
+- [PASS/WARN] Organization: <assessment>
+
+### Suggestions
+1. <most important improvement>
+2. <second improvement>
+3. <third improvement>
+```
+
+Be constructive. Focus on actionable improvements that will make the skill trigger more reliably and produce better outputs.

.claude/settings.json

@@ -0,0 +1,26 @@
+{
+  "hooks": {
+    "PreToolUse": [
+      {
+        "matcher": "Edit|Write",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "file_path=$(echo \"$CLAUDE_TOOL_INPUT\" | python3 -c \"import sys,json; print(json.load(sys.stdin).get('file_path',''))\" 2>/dev/null); if echo \"$file_path\" | grep -q 'marketplace.json$'; then echo 'BLOCK: marketplace.json should be updated via the plugin registration process, not edited directly. Follow docs/contributing.md or use the skill-template-generator skill.' >&2; exit 2; fi"
+          }
+        ]
+      }
+    ],
+    "PostToolUse": [
+      {
+        "matcher": "Edit|Write",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "file_path=$(echo \"$CLAUDE_TOOL_INPUT\" | python3 -c \"import sys,json; print(json.load(sys.stdin).get('file_path',''))\" 2>/dev/null); if echo \"$file_path\" | grep -q 'SKILL.md$'; then if [ -f \"$file_path\" ]; then frontmatter=$(sed -n '/^---$/,/^---$/p' \"$file_path\" | head -30); if ! echo \"$frontmatter\" | grep -q '^name:'; then echo 'WARNING: SKILL.md is missing required name: field in frontmatter'; fi; if ! echo \"$frontmatter\" | grep -q '^description:'; then echo 'WARNING: SKILL.md is missing required description: field in frontmatter'; fi; fi; fi"
+          }
+        ]
+      }
+    ]
+  }
+}

.mcp.json

@@ -0,0 +1,15 @@
+{
+  "mcpServers": {
+    "context7": {
+      "command": "npx",
+      "args": ["-y", "@upstash/context7-mcp@latest"]
+    },
+    "github": {
+      "command": "npx",
+      "args": ["-y", "@modelcontextprotocol/server-github"],
+      "env": {
+        "GITHUB_PERSONAL_ACCESS_TOKEN": ""
+      }
+    }
+  }
+}

plugins/business-ops/skills/onboarding-checklist/SKILL.md

@@ -0,0 +1,95 @@
+---
+name: onboarding-checklist
+description: >
+  Generate onboarding checklists and guides for new team members at Rhombus.
+  Use this skill when the user asks to create an onboarding plan, new hire
+  checklist, onboarding guide, first-week plan, or team onboarding document.
+  Also trigger on: new employee setup, onboarding doc, welcome guide, getting
+  started guide, new hire orientation, onboarding steps, team setup checklist,
+  day one plan, 30-60-90 plan.
+disable-model-invocation: true
+argument-hint: "[role or team name]"
+---
+
+# Onboarding Checklist Generator
+
+Create structured onboarding checklists tailored to a specific role or team.
+
+## Information Gathering
+
+Ask the user for:
+
+1. **Role/title** — What position is being onboarded?
+2. **Team** — Which team are they joining?
+3. **Start date** (optional) — To set timeline milestones
+4. **Tools & systems** — What tools does this role use daily?
+5. **Key contacts** — Who should the new hire meet?
+6. **Special requirements** — Any certifications, access levels, or training needed?
+
+## Checklist Format
+
+```markdown
+# Onboarding Checklist: <Role> — <Team>
+**Created:** <date>
+**New Hire:** _________________
+**Manager:** _________________
+**Buddy:** _________________
+
+---
+
+## Pre-Start (Before Day 1)
+- [ ] Send welcome email with start date, time, and location
+- [ ] Set up accounts: <list relevant tools>
+- [ ] Order equipment: <laptop, monitors, etc.>
+- [ ] Add to team channels: <Slack channels>
+- [ ] Schedule Day 1 orientation meetings
+- [ ] Assign onboarding buddy
+
+## Week 1: Orientation & Setup
+### Day 1
+- [ ] Welcome meeting with manager
+- [ ] IT setup and account verification
+- [ ] Office/building tour (or remote setup walkthrough)
+- [ ] Review team norms and communication preferences
+- [ ] Meet onboarding buddy
+
+### Days 2-5
+- [ ] Complete required HR training
+- [ ] Read team documentation and READMEs
+- [ ] Set up development environment (if engineering)
+- [ ] Attend team standup/sync
+- [ ] 1:1 meetings with key collaborators
+- [ ] Complete first small task or shadowing session
+
+## Weeks 2-4: Ramp Up
+- [ ] Take ownership of first real task/project
+- [ ] Attend all recurring team meetings
+- [ ] Complete role-specific training: <list>
+- [ ] Review and understand team OKRs/goals
+- [ ] Weekly 1:1 with manager (ongoing)
+- [ ] Meet cross-functional partners
+
+## Month 2-3: Full Contribution
+- [ ] Own a workstream independently
+- [ ] Contribute to planning/sprint sessions
+- [ ] 30-day check-in with manager
+- [ ] 60-day check-in with manager
+- [ ] 90-day review and goal-setting
+
+---
+
+## Key Resources
+| Resource | Link |
+|----------|------|
+| <handbook/wiki> | <url> |
+| <tool docs> | <url> |
+| <team page> | <url> |
+```
+
+## Guidelines
+
+- Tailor the checklist to the specific role — an engineer's onboarding differs from a marketer's
+- Include both administrative tasks (accounts, HR) and social tasks (meet the team)
+- Set realistic timelines — don't expect full productivity in week 1
+- Include checkboxes so the document is actionable
+- Note who is responsible for each item (manager, IT, buddy, new hire)

plugins/business-ops/skills/status-report/SKILL.md

@@ -0,0 +1,76 @@
+---
+name: status-report
+description: >
+  Generate structured status reports, project updates, and weekly summaries
+  for leadership and stakeholders. Use this skill when the user asks to write
+  a status report, project update, weekly update, progress report, leadership
+  update, or team summary. Also trigger on: status update, standup summary,
+  executive summary, progress check, what's the status, weekly recap, sprint
+  summary, project status, team update, 3P update, stakeholder update.
+disable-model-invocation: true
+argument-hint: "[project or team name]"
+---
+
+# Status Report Generator
+
+Generate clear, structured status reports for leadership and stakeholders.
+
+## Information Gathering
+
+Before writing the report, ask the user for:
+
+1. **Reporting period** — What timeframe does this cover? (e.g., this week, this sprint, this month)
+2. **Audience** — Who is this for? (e.g., leadership, cross-functional partners, the team)
+3. **Key accomplishments** — What was completed or shipped?
+4. **In progress** — What's currently being worked on?
+5. **Blockers/risks** — Anything slowing the team down or at risk?
+6. **Upcoming** — What's planned for the next period?
+7. **Metrics** (optional) — Any numbers worth highlighting?
+
+If the user provides context (e.g., git log, Jira/Linear data, meeting notes), use it to populate the report instead of asking.
+
+## Report Format
+
+```markdown
+# Status Report: <Project/Team Name>
+**Period:** <date range>
+**Author:** <name>
+**Date:** <today>
+
+---
+
+## Summary
+<2-3 sentence executive summary of the most important takeaway>
+
+## Completed
+- <accomplishment with context on impact>
+- <accomplishment>
+
+## In Progress
+- <work item> — <current status, % complete or ETA>
+- <work item> — <status>
+
+## Blockers & Risks
+- [BLOCKER] <description> — **Owner:** <who can unblock> — **Impact:** <what's affected>
+- [RISK] <description> — **Likelihood:** High/Med/Low — **Mitigation:** <plan>
+
+## Upcoming (Next Period)
+- <planned work item>
+- <planned work item>
+
+## Metrics (if applicable)
+| Metric | Previous | Current | Trend |
+|--------|----------|---------|-------|
+| <metric> | <value> | <value> | <up/down/flat> |
+
+## Asks / Decisions Needed
+- <any decisions or approvals needed from the audience>
+```
+
+## Writing Guidelines
+
+- Lead with impact, not activity — "Shipped feature X, reducing support tickets by 30%" not "Worked on feature X"
+- Keep each bullet to 1-2 lines — this is a scan document, not a narrative
+- Be honest about blockers — surfacing problems early builds trust
+- Quantify when possible — numbers are more memorable than adjectives
+- Match the formality to the audience — leadership gets concise summaries, team gets more detail