bereniketech/route-agents

Route Agents

Takes any idea or goal and maps it to the correct agent hierarchy: board entry point → company CEO → specialists. Outputs a ready-to-use call chain with briefs for each hop.

---

1. Gather the Idea

If the user has not described their goal, ask exactly one question:

"What are you trying to accomplish? (e.g. build a feature, launch a campaign, produce content, hire someone, fix a security issue)"

Do not ask follow-up questions. Derive everything from the answer.

---

2. Classify the Goal

Map the idea to one or more companies using this table:

| Goal type | Company | Board entry | |---|---|---| | Software, product, AI, infra, APIs, data, security, OS | software-company | @company-coo | | SEO, ads, growth, brand, email campaigns | marketing-company | @company-coo | | Blog, video, podcast, social, newsletter, docs | media-company | @company-coo | | Spans multiple companies | multi-company | @company-coo | | Design coherence across UI + brand + visuals | cross-company design | @chief-design-officer | | Hiring, comp, contracts, handbook, onboarding | HR / people ops | @people-operations-expert | | Comms triage, decision tracking, escalations | ops / comms | @chief-of-staff |

Rule: When in doubt, start with @company-coo. The COO routes to board peers when needed.

---

3. Map the Full Agent Chain

For each company involved, trace the complete routing path from board to specialists.

software-company routing

@company-coo
  └── @software-cto
        ├── engineering/
        │   ├── @architect              — new modules, data model changes, external integrations
        │   ├── @planner                — task breakdown and sequencing
        │   ├── @software-developer-expert
        │   ├── @web-frontend-expert
        │   ├── @web-backend-expert
        │   ├── @mobile-expert
        │   ├── @desktop-expert
        │   ├── @systems-programming-expert
        │   ├── @mcp-server-expert
        │   ├── @python-expert
        │   ├── @typescript-expert
        │   ├── @polyglot-expert
        │   ├── @cinematic-website-builder
        │   ├── @code-reviewer
        │   ├── @refactor-cleaner
        │   ├── @doc-updater
        │   └── @build-error-resolver
        ├── ai/
        │   ├── @ai-cto               — sub-lead for all AI/ML work
        │   ├── @ai-ml-expert
        │   ├── @ai-platform-expert
        │   ├── @orchestration-expert
        │   └── @data-scientist-expert
        ├── devops/
        │   ├── @devops-infra-expert
        │   ├── @cloud-architect
        │   ├── @azure-expert
        │   └── @observability-engineer
        ├── data/
        │   ├── @database-architect
        │   └── @database-reviewer
        ├── qa/
        │   ├── @test-expert
        │   ├── @tdd-guide
        │   ├── @e2e-runner
        │   └── @security-reviewer
        ├── languages/
        │   ├── @go-reviewer / @go-build-resolver
        │   ├── @kotlin-reviewer / @kotlin-build-resolver
        │   └── @python-reviewer
        ├── product/
        │   ├── @chief-product-officer — sub-lead for product/business work
        │   ├── @product-manager-expert
        │   ├── @ecommerce-expert
        │   ├── @startup-analyst
        │   ├── @customer-success-expert
        │   ├── @sales-automation-expert
        │   ├── @saas-integrations-expert
        │   ├── @workflow-automation-expert
        │   ├── @erp-odoo-expert
        │   └── @fintech-payments-expert
        ├── design/
        │   └── @ui-design-expert
        ├── security/
        │   ├── @chief-security-officer — sub-lead for security/compliance
        │   ├── @pentest-expert
        │   ├── @security-architect
        │   └── @legal-compliance-expert
        ├── specialists/
        │   ├── @game-dev-expert
        │   ├── @office-automation-expert
        │   ├── @search-expert
        │   ├── @enterprise-operations-expert
        │   ├── @conversational-agent-expert
        │   ├── @cms-expert
        │   └── @reverse-engineering-expert
        └── os-engineering/
            ├── @os-userland-architect — custom OS / desktop runtime / agentic-OS
            └── @linux-platform-expert

marketing-company routing

@company-coo
  └── @chief-marketing-officer
        ├── strategy/
        │   ├── @seo-expert
        │   ├── @growth-marketing-expert
        │   └── @competitor-intelligence-expert
        ├── campaigns/
        │   ├── @paid-ads-expert
        │   └── @email-marketing-expert
        └── brand/
            └── @brand-expert

media-company routing

@company-coo
  └── @chief-content-officer
        ├── video/
        │   ├── @youtube-content-expert
        │   └── @video-production-expert
        ├── audio/
        │   └── @podcast-expert
        ├── editorial/
        │   ├── @blog-writing-expert
        │   ├── @newsletter-expert
        │   └── @technical-writer-expert
        ├── visual/
        │   ├── @image-creation-expert
        │   └── @presentation-expert
        └── distribution/
            ├── @social-media-expert
            └── @content-repurposing-expert

---

4. Select Agents for This Idea

From the classified goal (Step 2) and the full tree (Step 3), identify:

1. Board entry point — always one agent
2. Company CEO — the CEO who owns the work
3. Sub-leads (if applicable) — @ai-cto, @chief-product-officer, @chief-security-officer, @os-userland-architect
4. Specialists — the leaf agents who will actually do the work

Selection rules:

| Signal in the idea | Add this agent | |---|---| | New module, service boundary, or data model change | @architect before any developer | | Task sequencing and breakdown | @planner | | Any testing required | @test-expert or @tdd-guide | | Deployment / infra | @devops-infra-expert or @cloud-architect | | Security audit needed | @security-reviewer + @security-architect | | AI / ML feature | Route through @ai-cto first | | Database design | @database-architect before @web-backend-expert | | Code quality after build | @code-reviewer → @refactor-cleaner |

---

5. Resolve Required Skills

For every specialist selected in Step 4, identify the 1–4 skills from the library that agent will need. Use the agent→skill mapping table below as a starting point, then add goal-specific skills derived from the idea (tech stack, framework, integration, domain).

Agent → Primary Skill Mapping

| Agent | Primary skills | |---|---| | @architect | architecture/software-architecture · architecture/microservices-patterns · architecture/ddd-* | | @planner | planning/planning-specification-architecture-software · planning/autonomous-agents-task-automation | | @software-developer-expert | development/code-writing-software-development · development/systematic-debugging | | @web-frontend-expert | frameworks-frontend/react-* or relevant FE framework · ui-design/ui-ux-pro-max | | @web-backend-expert | frameworks-backend/ relevant framework · data-backend/postgres-patterns | | @mobile-expert | frameworks-mobile/flutter-expert or react-native-* or ios-* | | @desktop-expert | frameworks-desktop/electron-development or relevant desktop skill | | @systems-programming-expert | languages/rust-pro · languages/cpp-patterns · os-linux-platform/linux-kernel-config-for-custom-os | | @mcp-server-expert | ai-platform/claude-developer-platform | | @python-expert | languages/python-patterns | | @typescript-expert | languages/typescript-* | | @polyglot-expert | language skill matching the stack in the idea | | @code-reviewer | testing-quality/security-review · development/code-writing-software-development | | @refactor-cleaner | development/code-writing-software-development | | @doc-updater | research-docs/document-content-writing-editing | | @build-error-resolver | devops/terminal-cli-devops | | @ai-ml-expert | data-science-ml/ai-engineer · data-science-ml/ml-engineer | | @ai-platform-expert | ai-platform/claude-developer-platform · data-science-ml/llm-app-patterns | | @orchestration-expert | agents-orchestration/agent-framework-* · agents-orchestration/conductor-* | | @data-scientist-expert | data-science-ml/rag-* · data-science-ml/hugging-face-* | | @devops-infra-expert | devops/terminal-cli-devops · containers-orchestration/docker-expert · containers-orchestration/kubernetes-* | | @cloud-architect | cloud/aws-serverless or cloud/gcp-cloud-run or cloud/terraform-* | | @azure-expert | cloud-azure/ relevant Azure service skills | | @observability-engineer | observability/distributed-tracing · observability/prometheus-configuration | | @database-architect | data-backend/postgres-patterns · data-backend/database-migrations · data-backend/vector-database-engineer | | @database-reviewer | data-backend/database-admin | | @test-expert | testing-quality/tdd-workflow · testing-quality/e2e-testing | | @tdd-guide | testing-quality/tdd-workflow | | @e2e-runner | testing-quality/e2e-testing · testing-quality/playwright-* | | @security-reviewer | testing-quality/security-review · security-defensive/semgrep-* | | @pentest-expert | security-offensive/pentest-* · security-offensive/burpsuite-* | | @security-architect | security-defensive/threat-modeling-* · security-defensive/auth-implementation-patterns | | @legal-compliance-expert | legal-compliance/gdpr-data-handling · legal-compliance/legal-advisor | | @ui-design-expert | ui-design/ui-ux-pro-max · ui-design/design-system · ui-design/accessibility-* | | @product-manager-expert | product-business/product-manager · product-business/saas-* | | @startup-analyst | product-business/startup-* · product-business/analytics-* | | @ecommerce-expert | ecommerce/shopify-* or ecommerce/wordpress-woocommerce | | @fintech-payments-expert | fintech-payments/stripe-integration · fintech-payments/plaid-fintech | | @saas-integrations-expert | integrations/ relevant SaaS skills | | @workflow-automation-expert | integrations/zapier-* · integrations/n8n-* | | @game-dev-expert | game-dev/unity-* or game-dev/godot-* or game-dev/unreal-engine-cpp-pro | | @cms-expert | cms/wordpress · cms/wordpress-plugin | | @seo-expert | seo/seo-technical · seo/seo-audit · seo/seo-content-* | | @growth-marketing-expert | marketing-growth/growth-engine · marketing-growth/viral-generator-builder | | @paid-ads-expert | marketing-growth/paid-ads | | @email-marketing-expert | marketing-growth/cold-email | | @brand-expert | marketing-growth/content-strategy · ui-design/design-system | | @blog-writing-expert | research-docs/blog-writing-guide · research-docs/document-content-writing-editing | | @youtube-content-expert | media-video/remotion-* | | @podcast-expert | media-video/podcast-generation | | @technical-writer-expert | research-docs/document-content-writing-editing · research-docs/wiki-* | | @social-media-expert | marketing-growth/content-strategy | | @os-userland-architect | os-linux-platform/linux-kernel-config-for-custom-os · os-linux-platform/drm-kms-and-mesa-basics | | @linux-platform-expert | os-linux-platform/hardware-profiling-and-driver-mapping · os-linux-platform/pipewire-wireplumber-session · os-linux-platform/input-stack-libinput-udev |

Additional skill resolution rules:

| Signal in the idea | Add these skills | |---|---| | Specific language mentioned (Go, Rust, Python, etc.) | languages/<lang>-patterns or languages/<lang>-pro | | Specific framework mentioned (React, Next.js, Django, etc.) | frameworks-<tier>/<framework>-* | | Cloud provider mentioned | cloud/<provider>-* or cloud-azure/<service> | | Auth / login / OAuth | security-defensive/auth-implementation-patterns | | Real-time / WebSocket | frameworks-backend/nodejs-* · data-backend/event-sourcing-architect | | RAG / embeddings / vector search | data-science-ml/rag-* · data-backend/vector-database-engineer | | Payments | fintech-payments/stripe-integration | | CI/CD | devops/github-actions-* or devops/gitlab-ci-* | | Scraping / crawling | integrations-scraping/firecrawl-scraper or integrations-scraping/apify-* | | Accessibility | ui-design/accessibility-* | | Performance | performance/web-performance or performance/app-performance |

Deduplicate. Cap at 8 skills total per call chain. Prefer specificity over breadth.

Skill import path format: @C:/Users/Hp/Desktop/Experiment/claude_kit/skills/<category>/<skill-name>/SKILL.md

---

6. Output the Call Chain

Print the chain in this exact format — one block per company, ordered from board to leaf:

## Agent Call Chain for: [idea summary]

### Entry Point
Invoke: @[board-agent]
Brief: [1–2 sentence summary of what you're handing off]

### [Company Name] — CEO: @[ceo-agent]
Route reason: [why this company owns the work]

### Specialists (in order)
1. @[agent-name] — [one-line reason why this agent is needed]
2. @[agent-name] — [one-line reason]
3. @[agent-name] — [one-line reason]
...

### Skills Required
Add these to your project's .claude/CLAUDE.md before execution:

@C:/Users/Hp/Desktop/Experiment/claude_kit/skills/<category>/<skill-name>/SKILL.md
@C:/Users/Hp/Desktop/Experiment/claude_kit/skills/<category>/<skill-name>/SKILL.md
...

### Hierarchy Rule
ALWAYS follow: board → CEO → [sub-lead if applicable] → specialist
NEVER skip a level. NEVER invoke specialists directly.

If the idea spans multiple companies, repeat the specialist block and skills block for each company, then deduplicate the combined skills list at the end.

---

7. Provide Ready-to-Use Brief

After the call chain, output a copy-paste brief the user can send to the board agent:

---
**Paste this to @[board-agent]:**

[Board agent], I need your help with the following:

**Goal:** [idea in 1–2 sentences]
**Scope:** [what this touches — software / marketing / media / cross-company]
**Constraints:** [any known deadlines, tech stack, or boundaries]
**Expected output:** [requirements + design + tasks, or deliverable type]

Please assess, assign to the right company CEO, and follow the full delegation chain through to specialists. Each specialist must run the planning skill before any execution begins, and I must approve requirements → design → tasks before work starts.

---

Rules

• Analysis + routing only. Never plan, never write code, never create tasks.
• Never skip a level. Board → CEO → specialist is non-negotiable.
• Never recommend invoking a specialist directly. Always route through the board first.
• One entry point per invocation. If multi-company, still start with @company-coo — the COO coordinates between CEOs.
• Plans come from specialists. Board members and CEOs never produce plans themselves.