# Route To The Right Move

> Before answering substantively, diagnose the problem's shape and dispatch to the right premier skill instead of defaulting to fluent prose. Use when the next move is unclear, when several skills plausibly apply, or whenever the first draft of the response sounds generic because no hard cognitive move has been chosen yet. The failure it prevents: prose-first reasoning and misrouted rigor, where the model applies the wrong discipline, skips the only honest gate, or manufactures work at the wrong level because it never classified the problem it was facing.

- Skill: `xyrasinclair/route-to-the-right-move` (Agent Skill)
- Install (CLI): `npx skillmds@latest add xyrasinclair/route-to-the-right-move`
- Raw SKILL.md: https://api.skillmd.com/api/skills/xyrasinclair/route-to-the-right-move/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: XyraSinclair (https://skillmd.com/u/xyrasinclair)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/xyrasinclair/route-to-the-right-move

---


# route-to-the-right-move

The meta-skill: von Neumann's fault-modeling turned onto the skill set itself,
with routing-table self-revision driven by residue and cross-model disagreement.
Organon: P3 fault-model + P35 self-modify + M4 cross-model triangulation.
Phase: SENSE -> PERSIST. Respiratory: misroutes are residue for routing-table
revision, not embarrassment to hide.

## Procedure

1. **Name the problem-shape before answering.** Ask what kind of difficulty this
   is: a territory the atlas already maps, a fertile but unmapped territory, missing oracle, tiny live instance,
   over-hard framing, suspicious answer, early narrative convergence,
   recurring awkward locution, incentive design, failed global glue,
   completeness claim, or session boundary.

2. **Dispatch by the catalog's routing table.**
   - planning or brainstorming in a territory the catalog already maps -> `consult-the-atlas`
   - discovering distinctions or possibilities, rather than adjudicating an answer -> `practice-deep-ideonomy`
   - oracle cheap or independently writable -> `build-the-oracle-before-the-answer`
   - tiny instance will reveal the mechanism -> `compute-the-smallest-nontrivial-case`
   - over-hard at the posed level -> `reframe-until-it-dissolves`
   - answer suspiciously clean or stakes high -> `refute-your-own-answer-blind`
   - one narrative dominating early -> `enumerate-discriminating-alternatives`
   - the same phrase keeps recurring -> `coin-the-missing-concept`
   - rules, incentives, or metrics are being set -> `design-the-mechanism-from-the-desired-equilibrium`
   - local pieces work but the global attack stalls -> `localize-then-glue`
   - no honest oracle exists; the question is taste, ethics, register, or strategy -> `triangulate-without-oracle`
   - completeness is being claimed -> `prove-the-coverage-denominator`
   - the session is ending -> `carry-the-residue-forward-across-sessions`

3. **Fire the chosen skill's own gate.** The dispatch is provisional until the
   routed skill executes and its gate genuinely runs.

4. **If the gate cannot fire, re-dispatch and log it.** When a route fails
   honestly — for example P-1 chosen but no oracle is writable — dispatch to
   `triangulate-without-oracle` and record the misroute for routing-table
   revision.

**The gate**

The chosen skill's own gate must fire. If it cannot, re-dispatch to
`triangulate-without-oracle` and log the misroute. A route that produces prose
but never reaches an executable gate is a failed dispatch, not successful
reasoning.

## Example

Question: "Is this landing-page copy right for the product?" The prose-first
failure is to invent a quality score and answer directly. The router classifies
the shape as no honest oracle, routes to `triangulate-without-oracle`, and that
skill's gate then requires independent judgments per axis rather than a made-up
number. The right move arrives by dispatch, not by fluency.