Skip to content

Guild-Agents/guild / src/templates/skills/re-specialize

re-specialize

Verified

Incremental re-specialization — re-scans the project and updates only auto-generated zones in CLAUDE.md and agents

FileKindSize
SKILL.mdskill md5.0 KB
evals/evals.jsonconfig1.9 KB
evals/triggers.jsonconfig842 B

evals/evals.json

View on GitHub
{
  "skill": "re-specialize",
  "evals": [
    {
      "id": "rs-has-core-steps",
      "description": "Re-specialize has read, explore, check-zones, regenerate, update, confirm, commit steps",
      "expectations": [
        { "text": "Has read-current step", "assertion": "step-exists:read-current" },
        { "text": "Has explore-project step", "assertion": "step-exists:explore-project" },
        { "text": "Has check-zones step", "assertion": "step-exists:check-zones" },
        { "text": "Has regenerate-zones step", "assertion": "step-exists:regenerate-zones" },
        { "text": "Has update-agents step", "assertion": "step-exists:update-agents" },
        { "text": "Has confirm step", "assertion": "step-exists:confirm" },
        { "text": "Has commit step", "assertion": "step-exists:commit" }
      ]
    },
    {
      "id": "rs-reasoning-for-regeneration",
      "description": "Zone regeneration uses reasoning tier (opus)",
      "expectations": [
        { "text": "regenerate-zones uses reasoning", "assertion": "step-model-tier:regenerate-zones:reasoning" }
      ]
    },
    {
      "id": "rs-execution-for-agents",
      "description": "Agent updates use execution tier (sonnet)",
      "expectations": [
        { "text": "update-agents uses execution", "assertion": "step-model-tier:update-agents:execution" }
      ]
    },
    {
      "id": "rs-gates",
      "description": "Gates at exploration, zone check, and confirmation",
      "expectations": [
        { "text": "explore-project has gate", "assertion": "gate-exists:explore-project" },
        { "text": "check-zones has gate", "assertion": "gate-exists:check-zones" },
        { "text": "confirm has gate", "assertion": "gate-exists:confirm" }
      ]
    },
    {
      "id": "rs-minimum-steps",
      "description": "Has at least 7 steps",
      "expectations": [
        { "text": "At least 7 steps", "assertion": "step-count:7" }
      ]
    }
  ]
}
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: 471kUpdated Claude CodeCodexVerifiedInstall
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