Gear-Sets (deterministic loadouts per loop type) · gear-sets
Owner directive (2026-07-10), canonical. "…setting the neuralware, the gear setup of the dynamic squad and each optimal suite of context, rules, models, resources, sparks between loops, ventures, endeavors, playbooks, cookbooks, blueprints, schematics, schemas, references, benchmark which allow the datan to be a deterministic agent to complete the task, since it may just change its neural ware to find the suitable resource for the development particles it shall looping in alongside its squad to augment." [src: wiki/development/prompts-archive/by-topic/knowledge-management-and-concept-modeling.md:453]
Content
A gear-set is a named, closed loadout — cookbooks + benchmark gates + schemas/SOPs + registry neuralwares (neuralware-index) — bound to a loop type for this project. Determinism rule: every playbook row names exactly one gear-set, and the datan executing that row equips exactly that set (sop-001-loop-execution step 2). No improvised tooling; swaps happen only via the academy rule, within the same family.
The gear-sets of the HOS-Instance build
| Gear-set | Family | For loop type | Library cards equipped | Registry neuralwares | Gate |
|---|---|---|---|---|---|
splat-authoring | authoring | holonprint / gaussian-splatting scenery loops | cb-spark · cb-threejs · cb-patterns-royal-orbit | nw-code-writer, nw-studio, nw-devtools-149 | bm-002-splat-budget |
page-authoring | authoring | page/route composition loops | cb-react · cb-nextjs · cb-tailwind · cb-patterns-hosweb | nw-code-writer, nw-modern-web-guidance | bm-001-web-perf |
scroll-motion | authoring | scroll-film, transition and behavior loops | cb-lenis · cb-gsap · cb-patterns-royal-orbit | nw-code-writer, nw-devtools-149 | bm-001-web-perf (scroll long-task gate) |
build-tooling | tooling | build/CI/perf-instrumentation loops | cb-vite · bp-001-spatial-site | nw-build-runner, nw-git-ops, nw-devtools-149 | bm-001-web-perf (bundle gates) |
library-authoring | governance | knowledge-card authoring loops | sop-004-library-authoring · schema-library-json · sop-002-conflict-resolution | nw-palantir-ontology, nw-hieroglyph, nw-version-control | zero-orphan gate (sop-004-library-authoring §8) |
run-orchestration | governance | dispatch/heartbeat/leadership loops | sop-003-run-dispatch · schema-dispatch-json · pb-003-run-dispatch | nw-neural-engine, nw-loop-coordinator, nw-version-control | tick contract (sop-003-run-dispatch) |
Families and the swap rule
Families: authoring (splat-authoring, page-authoring, scroll-motion), tooling (build-tooling), governance (library-authoring, run-orchestration).
On a failed delivery the Datan "should go to the academy and bring a new neuralware (…) to try again" [src: …/knowledge-management-and-concept-modeling.md:612]. The replacement must come from the same family's pool (e.g. swap nw-studio → nw-visual-studio inside authoring); crossing families requires RUN Leadership re-dispatch (squad-roles). Every swap is heartbeat-logged (sop-003-run-dispatch).
Assignment
Default gear-sets per seat are declared in each squad card (squad-eggtech, squad-schematic, squad-instance); per-row overrides live in the playbook row itself (pb-001-spatial-site-loop, pb-002-library-authoring, pb-003-run-dispatch). A row without a gear-set fails the dispatch capability check (sop-003-run-dispatch step 2).
Composition
- uses: neuralware-index (the pool of equippable registry entries) · cb-spark / cb-threejs / cb-lenis / cb-gsap / cb-vite / cb-react / cb-nextjs / cb-tailwind (cookbooks packed into sets) · bm-001-web-perf / bm-002-splat-budget (gates packed into sets)
- used-by: sop-001-loop-execution (equip + swap semantics) · sop-003-run-dispatch (capability check) · squad-eggtech · squad-schematic · squad-instance (seat defaults) · okrs-2026-h2 (activity rows)
See also
- cognitive-loop — where an equipped neuralware injects into the sequence
- pb-002-library-authoring — the playbook whose rows equip
library-authoring