Skip to content

CruxExperts/LocalSetup / ls/skills/ls-agent-capability-routing

ls-agent-capability-routing

Verified

Use 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.

---
name: ls-agent-capability-routing
description: Use 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.
metadata:
  version: "1.0"
---

# LocalSetup Agent Routing

Use this skill when a bounded task needs a deterministic, reviewed `Agent-*`
lane recommendation. The selector reads only its bundled matrix and writes a
share-safe JSON receipt. It does not select a client model, effort, permission,
or worker configuration.

## Select a lane

Provide a closed JSON request on standard input or in a file:

```bash
python3 scripts/agent_routing.py select --request request.json
```

```bash
printf '%s' '{"schema":"agent_routing_request_v1","task_class":"routine","risk":"low","required_capabilities":[]}' \
  | python3 scripts/agent_routing.py select --request -
```

Treat `rejected` and `offline` receipts as typed results. Do not add a model,
price, account, observation, permission, or Ultra override to the request.
Request input is limited to 65,536 UTF-8 bytes; oversized, unavailable, or
malformed input is returned as a rejected `invalid_request` receipt after the
bundled resource passes validation. Resource validation runs first, so an invalid
or expired resource returns `resource_invalid` or `resource_stale` without
reading or classifying the request.

## Boundaries

- The matrix is static reviewed evidence. Unknown capability evidence never
  satisfies a request.
- `fresh_until` is the repository's local review-expiry policy, not an upstream
  freshness guarantee.
- `Agent-*` is a LocalSetup routing-policy label, not a provider model claim.
- The selector is read-only and does not inspect runtime availability or
  observations.
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: 472kUpdated Claude CodeCodexUnverifiedInstall
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: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
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 availableStars: 322kUpdated Claude CodeCodexVerifiedInstall