with one click
model-selection
Determines which LLM model to use for each agent spawn
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Menu
Determines which LLM model to use for each agent spawn
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Generate an executive-ready Migration Strategy Report (HTML deck) from a customer portfolio — CMDB exports, RVTools/Azure Migrate inventories, DMA outputs, vendor proposals, architecture diagrams. **Use when:** producing a CIO-ready migration plan for a customer portfolio (apps, databases, infra, or any mix). Starts the Portfolio Planning flow for multi-app engagements. **Triggers:** "generate migration report", "analyze CMDB", "portfolio analysis", "migration strategy report for X", "RVTools export analysis", "Azure Migrate analysis", "portfolio assessment", "build a deck from these artifacts". **Covers:** auto-detected workload pillars (Apps/DB/Infra), CAF-aligned 6 Rs classification, deterministic Factory / ISD-Partner / Unknown ownership, phased roadmap, dependency mapping, move groups, risks, executive HTML output.
Shifts model selection to cost-optimized alternatives when economy mode is active
Find and resume interrupted Copilot CLI sessions using session_store queries
Docker containerization patterns for .NET and Java applications targeting Azure. **Use when:** Creating Dockerfiles or deploying to Container Apps, App Service containers, or AKS. **Triggers on:** Container Apps or AKS selected as hosting platform, Docker-related requests. **Covers:** Multi-stage Dockerfiles, docker-compose, Container Apps configuration, health checks, resource limits.
Azure Infrastructure as Code patterns using Bicep and Terraform. **Use when:** Generating IaC for App Service, Container Apps, or AKS deployments. **Triggers on:** Phase 3 infrastructure generation, azd init, Bicep/Terraform file creation. **Covers:** Azure Verified Modules, managed identities, Key Vault with RBAC, Application Insights, networking.
Business logic discovery, mapping, and preservation during application migration. **Use when:** Migrating .NET or Java applications and need to ensure all business logic is preserved. **Triggers on:** Phase 2 code migration, business rule identification, logic preservation verification. **Covers:** Business rule extraction, logic mapping between source/target, validation tracking, media asset preservation. **Critical:** Ensures no business logic is lost during migration by creating traceable mappings.
| name | model-selection |
| description | Determines which LLM model to use for each agent spawn |
| domain | orchestration |
| confidence | medium |
| source | adapted from squad-sdk |
Determines which LLM model to use for each agent spawn.
✅ THIS SKILL PRODUCES:
model parameter for every task tool call.squad/config.json❌ THIS SKILL DOES NOT PRODUCE:
Resolution is first-match-wins — the highest layer with a value wins.
| Layer | Name | Source | Persistence |
|---|---|---|---|
| 0a | Per-Agent Config | .squad/config.json → agentModelOverrides.{name} | Persistent (survives sessions) |
| 0b | Global Config | .squad/config.json → defaultModel | Persistent (survives sessions) |
| 1 | Session Directive | User said "use X" in current session | Session-only |
| 2 | Charter Preference | Agent's charter.md → ## Model section | Persistent (in charter) |
| 3 | Task-Aware Auto | Code → sonnet, docs → haiku, architecture → opus | Computed per-spawn |
| 4 | Default | claude-sonnet-4.6 | Hardcoded fallback |
| Task Type | Model | Rationale |
|---|---|---|
| Code migration, refactoring | claude-sonnet-4.6 | Best balance of speed and code quality |
| Architecture, complex design | claude-opus-4.6 | Deep reasoning for migration planning |
| Documentation, walkthroughs | claude-haiku-4.5 | Fast, good enough for prose |
| Security review | claude-sonnet-4.6 | Needs precision for vulnerability detection |
| Infrastructure (Bicep/IaC) | claude-sonnet-4.6 | Template generation needs accuracy |
| Cost analysis | claude-haiku-4.5 | Structured output, lower complexity |
| Discovery & intake | claude-sonnet-4.6 | Classification needs broad knowledge |
| Presentation/PPTX | claude-haiku-4.5 | Slide generation is structured |
| Command | Effect |
|---|---|
| "use opus for architecture" | Sets Layer 1 override for architecture tasks |
| "always use sonnet" | Sets Layer 0b in .squad/config.json |
| "reset model preferences" | Clears Layer 0 overrides |