katopz
@katopz
Public Skills
boundary-guard
by katopz
Audit + enforce game-stack boundary rules across the multi-repo workspace. Use when adding a System impl, game logic, vocabulary type, FFI surface, or view-layer code; when reviewing PRs touching game systems or view/FFI boundaries; when a violation is suspected; or quarterly as a boundary-hygiene gate. Reads each repo's BOUNDARY.md as the contract — owns, not-owns, dep allowlist, drift ledger — enforced via ci_boundary_contract.sh (workspace dep graph + contract honesty, every repo with a root BOUNDARY.md, derived never counted) + ci_boundary_guard.sh (per-repo code logic) + grep checks. Sibling to feature-gate-audit + goat-audit + doc-sync.
doc-sync
by katopz
Synchronize each repo's .docs/ and README.md with recently succeeded plans/issues/benchmarks by diffing git history against the last documented entry. Use after landing a GOAT-passing plan, closing a batch of issues, or quarterly as a doc-hygiene gate. Covers every workspace repo and knows each repo's doc layout + where to record what, including each repo's root BOUNDARY.md contract (drift rows vs issue state).
substrate-first
by katopz
Pre-implementation DRY gate + existing-code drift audit for the multi-repo workspace. Use BEFORE writing any new System impl, trait, perception/cognition/emotion pipeline, state management, spatial query, or vocabulary type — to verify you're consuming existing substrate, not duplicating it. Also use to AUDIT existing code for parallel-system DRY violations (code that re-implements substrate under different names). The canonical defense is vocabulary translation; concepts ship under operator names (GenericSpatialBelief, decay_confidence), not English names ("threat field", "spatial hash") — a single-vocabulary grep returns ZERO hits even when the substrate fully exists. Sibling to boundary-guard + goat-audit + feature-gate-audit + doc-sync.
goat-audit
by katopz
Audit cross-repo GOAT/gain primitive cherry-pick status across the multi-repo stack (katgpt-rs upstream, riir- consumers). Detects stalls (default-on in katgpt-rs for ≥7 days with zero runtime wiring in riir-), DRY violations (duplicated substrate in riir-* that should consume katgpt-core), and SOLID violations. Use when auditing primitive cherry-pick coverage, before opening a plan that consumes a katgpt-rs primitive, or quarterly as a hygiene gate.
security-audit
by katopz
Security audit workflow for this Rust workspace — trust-boundary-first review with a coverage ledger, three-verdict findings (confirmed / needs_validation / rejected), and fix-verification gates. Use for security questions, focused reviews of wire, FFI, or secrets surfaces, or a full six-phase audit when the user asks for one.
research
by katopz
Research workflow for distilling ML/AI papers into modelless inference primitives, freeze/thaw runtime patterns, latent-space operations, AND model-based training plans across the multi-repo stack. Use when reading arxiv papers, deciding which repo a paper belongs in, creating .research/ notes or .plans/ files, implementing modelless inference primitives, or routing training-vs-inference insights. Enforces the commercial strategy (public engine / private runtime / private chain / private neuron-db / private training / private SDK facade / private product-domain / public inference-substrate riir-infer / public decision-serving riir-reflex), three-track system (modelless inference + self-adaptive runtime + model-based trained weights — all three exist across the stack, not just riir-train), latent-to-latent preference, and freeze/thaw-over-fine-tuning rule. Every task ends with a mandatory Claude verdict ping-pong (§5) before any file is committed.
proposal
by katopz
Write a reasoned architectural proposal (.proposals/NNN_*.md) grounded in focused codebase grep + prior-art paper search. Use when arguing for a design change, new primitive, or architectural decision — the question is "should we do X?" not "how do we do X?" (that's a plan) or "what does paper Y distill to?" (that's research). Enforces honest caveats, fusion lineage, and the existing proposal format. Searches arxiv + codebase + sibling-repo proposals/research/plans before writing.
feature-gate-audit
by katopz
Audit feature-gate status claims across the multi-repo stack. Use when a doc/plan/issue/commit-message claims a "promotion discrepancy", "not yet wired", or "Default-off until..." state, when fixing stale feature-gate comments, before promoting or demoting any feature flag, or quarterly as a feature-gate-hygiene gate. Enforces four defenses (1) source-code verification of every wiring claim — grep the production tick path, don't trust the doc; (2) multi-surface grep for stale comments across 5 documentation surfaces (source .rs, lib.rs module doc, Cargo.toml default block, downstream Cargo.toml, .benchmarks/*_promotion_review.md); (3) layer-split awareness — engine-layer DEFAULT-ON + game-layer OPT-IN is a deliberate pattern when each layer gates a different concern (generic runtime vs IP-bearing content), NOT a missed propagation; (4) gate-chain resolution — own cfg plus every ancestor mod plus the crate default list, settled by a build not a read. Sibling to goat-audit + doc-sync.
rust-optimize
by katopz
Optimize Rust code until nothing left to improve. Loops automatically. Use when the user says "optimize" or "/optimize".