CruxExperts/LocalSetup / ls/skills/ls-agent-capability-routing
ls-agent-capability-routing
VerifiedUse when choosing a LocalSetup Agent-* lane for exploration, research, bounded implementation, or review. Consult the static capability matrix; do not probe account or network availability.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 1.7 KB |
| resources/model-capability-matrix/manifest.json | config | 1.6 KB |
| resources/model-capability-matrix/schema.json | config | 2.9 KB |
| resources/model-capability-matrix/snapshot.json | config | 4.1 KB |
| schemas/routing-receipt.schema.json | config | 2.0 KB |
| schemas/routing-request.schema.json | config | 593 B |
| scripts/agent_routing.py | script | 27.7 KB |
resources/model-capability-matrix/snapshot.json
View on GitHub{
"$schema": "https://localsetup.dev/resources/model-capability-matrix/v1/schema.json",
"schema_version": 1,
"resource_id": "model-capability-matrix",
"owner_skill": "ls-agent-capability-routing",
"reviewed_at": "2026-11-02",
"fresh_until": "2026-12-02",
"candidates": [
{"lane": "Agent-Compact", "model_id": "gpt-6-luna", "eligibility": "reviewed", "policy_priority": 10, "task_classes": ["discovery", "routine", "research"], "maximum_risk": "medium", "capabilities": {"vision": {"value": "unknown"}}},
{"lane": "Agent-Realtime", "model_id": "gpt-6-luna", "eligibility": "reviewed", "policy_priority": 5, "task_classes": ["routine"], "maximum_risk": "low", "capabilities": {"vision": {"value": "unknown"}}},
{"lane": "Agent-Efficient", "model_id": "gpt-6.1-sol", "eligibility": "reviewed", "policy_priority": 30, "task_classes": ["routine", "implementation", "research", "review"], "maximum_risk": "high", "capabilities": {"vision": {"value": "unknown"}}},
{"lane": "Agent-Balanced", "model_id": "gpt-6.1-sol", "eligibility": "reviewed", "policy_priority": 20, "task_classes": ["implementation", "research", "review"], "maximum_risk": "high", "capabilities": {"vision": {"value": "unknown"}}},
{"lane": "Agent-Frontier", "model_id": "gpt-6-astra", "eligibility": "reviewed", "policy_priority": 50, "task_classes": ["discovery", "routine", "implementation", "research", "review"], "maximum_risk": "critical", "capabilities": {"vision": {"value": "unknown"}}}
],
"model_records": [
{"model_id": "gpt-6-astra", "evidence_id": "model-gpt-6-astra", "facts": {"page_label": "Default", "description": "Most capable model for the most demanding work."}},
{"model_id": "gpt-6.1-sol", "evidence_id": "model-gpt-6.1-sol", "facts": {"page_label": "Default", "description": "Near-Astra performance for complex work at a lower cost."}},
{"model_id": "gpt-6-luna", "evidence_id": "model-gpt-6-luna", "facts": {"page_label": "Default", "description": "Most efficient model for focused, high-volume tasks."}}
],
"scoped_records": [
{"record_id": "api-family-gpt-6", "scope": "api_family", "evidence_id": "api-family-gpt-6", "facts": {"model_reasoning_effort": {"gpt-6-astra": ["low", "medium", "high", "xhigh", "max"], "gpt-6.1-sol": ["low", "medium", "high", "xhigh", "max"], "gpt-6-luna": ["none", "low", "medium", "high", "xhigh", "max"]}, "input_modalities": ["text", "image"], "output_modalities": ["text"], "context_tokens": 1050000, "max_output_tokens": 128000, "endpoints": ["responses", "chat_completions", "batch"], "tools": ["web_search", "file_search", "image_generation", "code_interpreter", "hosted_shell", "apply_patch", "skills", "computer_use", "mcp", "tool_search"]}},
{"record_id": "codex-client-subagent-defaults", "scope": "client_product", "evidence_id": "client-codex-subagents", "facts": {"recommended_models": ["gpt-6.1-sol", "gpt-6-luna"], "documented_subagent_defaults": {"agents_enabled": true, "max_concurrent_threads_per_session": "client_selected_when_unset", "legacy_max_threads_alias": "agents.max_threads", "interrupt_message": true}}},
{"record_id": "api-pricing-standard-short-context", "scope": "accounting", "evidence_id": "accounting-api-pricing-standard-short-context", "facts": {"rate_class": "API Pricing Standard short-context", "unit": "USD_per_1M_tokens", "routing_use": "forbidden", "account_entitlement": "unknown", "plan_migration": "unknown", "models": {"gpt-6-astra": {"input": 10, "cached_input": 1, "cache_write": 12.5, "output": 50}, "gpt-6.1-sol": {"input": 2, "cached_input": 0.1, "cache_write": 2.5, "output": 10}, "gpt-6-luna": {"input": 0.1, "cached_input": 0.01, "cache_write": 0.125, "output": 0.5}}}},
{"record_id": "codex-token-rate-card", "scope": "accounting", "evidence_id": "accounting-codex-token-rate-card", "facts": {"rate_class": "Codex token rate-card", "unit": "credits_per_1M_tokens", "routing_use": "forbidden", "account_entitlement": "unknown", "plan_migration": "unknown", "models": {"gpt-6-astra": {"input": 250, "cached_input": 25, "output": 1250}, "gpt-6.1-sol": {"input": 50, "cached_input": 2.5, "output": 250}, "gpt-6-luna": {"input": 2.5, "cached_input": 0.25, "output": 12.5}}}}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |