LLMSkillHub
mirroredCopied from ruvnet/ruflo under MIT. ruvnet has not claimed this page, and is not affiliated with LLM SkillHub.

ruvnet / intelligence-route

Route tasks via the 3-tier model selector and learned patterns; emits a routing rationale via hooks_explain

0
SKILL.md
3.7 KB

Intelligence Routing

Pick the optimal agent + model tier for a task using learned patterns + the 3-tier router. Emits a hooks_explain rationale so the choice is auditable.

When to use

Before starting any non-trivial task. Replaces manual agent selection with data-driven decisions.

Steps

  1. Get an agent recommendation — mcp__plugin_ruflo-core_ruflo__hooks_route with the task description. Returns { recommended, confidence, reasoning }.
  2. Get a model tier recommendation — mcp__plugin_ruflo-core_ruflo__hooks_model-route for Haiku/Sonnet/Opus selection.
  3. Search for similar past patterns — mcp__plugin_ruflo-core_ruflo__hooks_intelligence_pattern-search to find prior successes.
  4. Predict outcome — mcp__plugin_ruflo-core_ruflo__neural_predict with the task description for a confidence-scored prediction.
  5. Spawn the recommended agent at the recommended model tier.
  6. (If --why was passed) — call mcp__plugin_ruflo-core_ruflo__hooks_explain to surface the routing rationale to the user.
  7. After task completes — call mcp__plugin_ruflo-core_ruflo__hooks_model-outcome with { task, model, outcome: "success"|"failure"|"escalated" } to train the router.

3-Tier Model Routing

TierHandlerLatencyCostWhen
1Deterministic codemod (TS compiler)~1ms$0Structural transforms with no LLM: var-to-const, remove-console, add-logging
2Haiku~500ms~$0.0002Low complexity (<30%), bug fixes, quick patches
3Sonnet/Opus2–5s$0.003–$0.015Complex reasoning, architecture, security, multi-file refactors

When hooks_route returns [CODEMOD_AVAILABLE] for a deterministic intent (var-to-const, remove-console, add-logging), call mcp__plugin_ruflo-core_ruflo__hooks_codemod with the intent + file — it applies the transform via the TypeScript compiler at $0, no LLM. Note: add-types, add-error-handling, async-await require judgement and route to a model (Tier 2/3) per ADR-143; they are NOT $0 codemods. Agent Booster is a fast-apply merge engine for LLM-produced edits, not the Tier-1 path.

Recording outcomes

Closing the routing loop is mandatory:

# Success (task, model and outcome are all required; model is haiku|sonnet|opus)
npx @claude-flow/cli@latest mcp exec -t hooks_model-outcome -p '{"task": "fix the flaky retry test", "model": "haiku", "outcome": "success"}'

# The model was too small and the work moved up a tier: use "escalated"; use "failure" if it did not work out
npx @claude-flow/cli@latest mcp exec -t hooks_model-outcome -p '{"task": "fix the flaky retry test", "model": "haiku", "outcome": "escalated"}'

The router learns from these calls. Skipping them = no learning.

CLI alternative

npx @claude-flow/cli@latest hooks route --task "description"
npx @claude-flow/cli@latest hooks pre-task --description "description"
npx @claude-flow/cli@latest hooks explain --task "description" --verbose