grtninja
- 157 skills
- 0 followers
- 1 week ago last updated
- ▌ Shim Pc Control Brain Routing · grtninja bundleEnforce one-brain routing across VRM Sandbox, PC Control, and MemryX shim with direct LM Studio model authority and typed MX3 support. Use when VRM chat/control behavior must be governed by PC Control without changing endpoint roles.
- ▌ Skill Arbiter Churn Forensics · grtninja bundleInvestigate rg.exe churn and quarantine decisions in skill-arbiter runs. Use when a skill causes CPU spikes, unexpected scans, blacklist actions, or when arbitration thresholds need evidence-backed tuning.
- ▌ Heterogeneous Stack Validation · grtninja bundleValidate a mixed local-plus-remote AI stack with privacy-safe evidence, endpoint health checks, inference smokes, and repo lockstep gates. Use when a request spans multiple local services, remote hosts, and heterogeneous inference lanes that must be tested end to end without launching restricted avatar endpoints.
- ▌ Skill Common Sense Engineering · grtninja bundleApply practical sanity checks before and after coding work to prevent avoidable mistakes. Use when verifying dependency surfaces, cleaning generated artifacts, checking privacy policy compliance, auditing artifact hygiene, or ensuring multi-skill chain guardrail evidence is present before claiming readiness.
- ▌ Skills Consolidation Architect · grtninja bundleConsolidate repository-specific skills into modular, reusable sets. Use when auditing skill overlap, splitting monolithic skills, reducing one-shot skills, defining per-repo core vs advanced skills, and planning safe deprecations with lockdown admission tests.
- ▌ Vrm Sandbox Startup Acceptance · grtninja bundleValidate and repair VRM Sandbox startup acceptance with shim-first local model authority, frontend/backend bring-up, and avatar-runtime launch proof. Use when launch behavior, chat handoff, voice fallback, or runtime bridge acceptance must be verified end to end.
- ▌ Vroid Vrma Photobooth Pipeline · grtninja bundleExtract VRoid Studio Photo Booth animation clips from a local AssetRipper Unity export and batch-convert them into valid .vrma files via Unity + UniVRM. Use when direct AssetBundle loading fails due Unity version/build-target mismatch.
- ▌ Meta Harness Four App Alignment · grtninja bundleAlign the four core local app surfaces under the shared meta-harness contract. Use when PC Control, shim/control-center, STARFRAME app, and avatar-runtime defaults or docs drift on model authority, canonical repo root, continuity, or endpoint expectations and one coordinated pass must repair them together.
- ▌ Repo B Local Comfy Orchestrator · grtninja bundleCompatibility wrapper for legacy local Comfy orchestration requests in <PRIVATE_REPO_B>. Route new MCP/Comfy operations to repo-b-mcp-comfy-bridge.
- ▌ Voice Catalog Runtime Alignment · grtninja bundleAlign documented voice-command catalogs, endpoint action allowances, and live runtime handlers so operator-visible voice surfaces match what the stack can actually execute. Use when voice command docs, parser matrices, endpoint permissions, or runtime action routing drift apart.
- ▌ Endpoint Admission Voice Actions · grtninja bundleAdmit Penny voice endpoints with explicit identity, role, and action limits before they can issue backend actions. Use when edge voice devices, endpoint profiles, or voice-initiated tool permissions change.
- ▌ Local Trace Evidence Correlation · grtninja bundleCorrelate local decision telemetry, trace recording, and emitted evidence across STARFRAME services. Use when arbiter, media, dashboard, or embodiment traces need to line up as one operator-readable storyline.
- ▌ Repo A Coordinator Smoke · grtninja bundleRun and debug Repo A DDC node/coordinator smoke workflows. Use when changing coordinator endpoints, job fetch/submit flow, runtime registration, or backend dispatch paths that must be verified with local coordinator execution.
- ▌ Repo C Ranking Contracts · grtninja bundleMaintain ranking engine behavior and schema compatibility in <PRIVATE_REPO_C>. Use when modifying scoring weights, ranking/multiplier logic, report serialization, or `ranking_report` schema contracts consumed by downstream systems.
- ▌ Repo D Setup Diagnostics · grtninja bundleDiagnose <PRIVATE_REPO_D> setup wizard and runtime state issues. Use when debugging LM Studio/Kokoro/Lovense setup paths, renderer hydration, profile persistence, overlay/background behavior, startup acceptance, or audio-baseline drift.
- ▌ Cybertron Fabric Host Ops · grtninja bundleAudit Cybertron remote-host readiness, repo parity, WinRM auth-state, MeshGPT host-terminal bootstrap, canonical model-matrix validation, Starframe/Cybertron offload checks, and VRM display readiness.
- ▌ Qwen Training Desktop Ops · grtninja bundleOperate the local Qwen Training Workbench desktop shell and its `9041` worker as one restart-safe local contract. Use when fixing or validating the Electron shell, launcher path, local dashboard load, queue or campaign controls, job visibility, or desktop startup acceptance for the private training app.
- ▌ Repo A Telemetry Kv Guard · grtninja bundleProtect Repo A DDC telemetry/privacy and encrypted KV pager contracts. Use when editing repo_c_trace adapters, Repo C policy gate logic, KV tiering/crypto code, or retention/privacy behavior tied to role acceptance.
- ▌ Repo B Control Center Ops · grtninja bundleOperate and debug <PRIVATE_REPO_B> Control Center and thin-waist service surfaces. Use when working on connector routing, Lighthouse checks, MCP/Agent Bridge endpoints, pose bridge, canonical desktop startup/restart behavior, shortcut ownership, or window lifecycle/state ownership.
- ▌ Repo B Preflight Doc Sync · grtninja bundleEnforce <PRIVATE_REPO_B> preflight gates and documentation lockstep. Use before PRs to run required validation profiles and keep README.md, docs/PROJECT_SCOPE.md, and docs/SCOPE_TRACKER.md synchronized with shipped behavior.
- ▌ Repo B Thin Waist Routing · grtninja bundleValidate and troubleshoot thin-waist REST routing in <PRIVATE_REPO_B>. Use when changing /v1 models/chat/vision routes, async job queue endpoints, connector routing policy, or bind/exposure behavior.
- ▌ Repo C Policy Schema Gate · grtninja bundleValidate Repo C policy files and schema contracts in <PRIVATE_REPO_C>. Use when editing policy manifests, schema files, policy validation CLI logic, or tests that rely on schema conformance.
- ▌ Skill Arbiter Release Ops · grtninja bundleRun release bump and PR release-hygiene workflow in skill-arbiter. Use when changes are release-impacting and require synchronized pyproject version, changelog entry, and CI release gate compliance.
- ▌ Skillhub Marketplace Sync · grtninja bundleCurate SkillHub marketplace discovery into repo-owned alignment artifacts, source reputation, and bounded promotion decisions. Use when SkillHub query sets, shortlist coverage, source ledgers, or alignment-matrix generation must be refreshed without promoting third-party skills past discovery-only by accident.
- ▌ Skills Discovery Curation · grtninja bundleDiscover, triage, and prioritize Codex skills for a repository or workspace. Use for one-time audits and recurring curation runs after cross-repo MX3/shim drift scans.
- ▌ Skills Third Party Intake · grtninja bundleVet third-party skill catalogs with deterministic security and quality scoring before arbiter admission. Use when mining external repos (for example openclaw/nullclaw) for safe candidate imports.
- ▌ Vroid Template Asset Sync · grtninja bundleDiscover and normalize VRoid/AvatarMaker clothing templates and texture assets into a deterministic manifest for Blender and Unity lanes. Use when template paths are inconsistent or assets were saved manually.
- ▌ Desktop Startup Acceptance · grtninja bundleEnforce explicit desktop startup acceptance across local app repos. Use when normalizing Windows app launchers, stale-window cleanup, frontend/backend coupling, state restore, shortcut ownership, or machine-specific app rosters.
- ▌ Playwright Edge Preference · grtninja bundleRun Playwright browser automation with Microsoft Edge as the default channel. Use when users request Edge-specific validation, screenshots, or parity checks while keeping deterministic low-churn automation flow.
- ▌ Qwen Training Campaign Ops · grtninja bundleOperate the Qwen Training Workbench campaign lane across `9041` worker queue ownership, named campaign catalogs, dependency-aware job expansion, and restart-safe continuation semantics. Use when campaign configs, worker APIs, or continuation scheduling change together.
- ▌ Repo B Agent Bridge Safety · grtninja bundleOperate and secure the Continue Agent Bridge in <PRIVATE_REPO_B>. Use when configuring bridge modes, validating /api/agent endpoints, enforcing controlled-write safety gates, or diagnosing bridge availability and permission failures.
- ▌ Repo C Boundary Governance · grtninja bundleKeep <PRIVATE_REPO_C> changes aligned with BOUNDARIES.md and AGENTS.md governance. Use when modifying trust-layer architecture, cross-repo interfaces, packaging docs, or any change that could blur cognitive vs hardware execution boundaries.
- ▌ Shockwave Operator Handoff · grtninja bundlePreserve Shockwave operator handoff across startup, control-rail contract display, chat/status surfaces, and voice/manual recovery behavior. Use when dashboard changes affect how the operator sees mode, authority, services, playback, or degraded-state transitions.
- ▌ Skill Cost Credit Governor · grtninja bundleGovern per-skill credit and token spend with deterministic warn/throttle/disable actions. Use when usage spikes, agent chatter, or budget overruns must be detected and contained.
- ▌ Media Workbench Desktop Ops · grtninja bundleOperate the STARFRAME Media Workbench desktop app and its local 9040 worker as one restart-safe local contract. Use when fixing or validating the Electron shell, source staging flow, prompt-head automation, queue submission, live telemetry, output preview, or desktop startup acceptance for the private media app.
- ▌ Operator Language Humanizer · grtninja bundleLocal-first operator language cleanup for prompts, run titles, queue notes, QC summaries, and user-visible output text across local app stacks. Use when tightening wording without changing meaning, policy, IDs, paths, flags, or system behavior. This skill is explicit opt-in and must never rewrite system prompts, personality files, or hidden instructions.
- ▌ Qwen Training Workbench Ops · grtninja bundleOperate the local teacher-student Qwen training workbench for long-running dataset-factory, staggered continuation batches, and training-agent supervision. Use when advancing or monitoring the private Huihui-to-Qwen3.5-4B pipeline, switching workstation modes, checking batch artifacts, or resuming overnight local training without media-workbench contention.
- ▌ Repo B Mx3 Router Contracts · grtninja bundleValidate MX3 shim model-router, ORT provider, and runtime packaging contracts. Use when changing route-mode derivation, router environment mapping, capability metadata, inference routing responses, exact ORT provider builds, or Electron/embedded-Node ABI ownership.
- ▌ Repo C Shim Contract Checks · grtninja bundleEnforce REPO_B shim runtime contract expectations in <PRIVATE_REPO_C>. Use when changing shim-facing adapters, telemetry dependencies, integration tests, or fail-closed behavior tied to REPO_B_SIDECAR_URL and mock-shim fixtures.
- ▌ Vrm Material Lighting Debug · grtninja bundleDiagnose and fix a desktop VRM app's avatar lighting and MToon material failures when a VRM looks washed out, has bright edges, shows inverted-looking cloth shadows, or shades clothing differently from skin and legs. Use when you need to inspect the live .vrm, separate scene-light problems from material problems, patch the correct layer, and verify with a rebuilt packaged app.
- ▌ Catalog Snapshot Consistency · grtninja bundleKeep generated catalogs, operator snapshots, and documented summaries aligned after workflow or command-surface changes. Use when visible command matrices, media/staging catalogs, or skill/status summaries drift from the runtime or generated source of truth.
- ▌ Media Staging Prompthead Ops · grtninja bundleOperate Media Workbench source staging, prompt-head generation order, lane assignment, and resume-safe draft continuity as one workflow surface. Use when staging, previews, or generation-order logic change together.
- ▌ Repo C Trace Ndjson Validate · grtninja bundleValidate Repo C/repo_c_trace NDJSON packet integrity in <PRIVATE_REPO_C>. Use when changing trace packet fields, guardian reroute behavior, or validation tooling that checks packet monotonicity and required envelope keys.
- ▌ Skill Blast Radius Simulator · grtninja bundleSimulate pre-install/pre-enable skill impact and require acknowledgement when risk exceeds thresholds. Use when admitting new skills or evaluating potentially risky updates.
- ▌ Repo B Wsl Hybrid Ops · grtninja bundleOperate and diagnose the Windows-host plus WSL-auxiliary split for <PRIVATE_REPO_B>. Use when validating hybrid service reachability, enforcing hardware-on-Windows boundaries, or debugging cross-host connectivity for ComfyUI and LM Studio helpers.
- ▌ Repo C Mass Index Ops · grtninja bundleRun shard-first mass-index operations for <PRIVATE_REPO_C> governance lanes. Use when scanning policy schemas, ranking/trace contracts, and trust-layer boundaries in very large trees where scope partitioning is required.
- ▌ Repo D Mass Index Ops · grtninja bundleRun UI-and-packaging mass-index operations for <PRIVATE_REPO_D> workspace trees. Use when scanning renderer components, Electron startup files, workspace packages, and packaging scripts while aggressively excluding generated build artifacts.
- ▌ Vrm Roundtrip CI Gate · grtninja bundleGate VRM importer/exporter changes using Blender, UniVRM, and VRM4U round-trip metrics. Use when PRs can affect bones, expressions, materials, or animation fidelity.
- ▌ Multitask Orchestrator · grtninja bundleSplit multi-lane requests into deterministic parallel workstreams and merge them with explicit evidence checks. Use when a request has 2+ independent objectives that can run concurrently.
- ▌ Repo D Local Packaging · grtninja bundleRun Windows-local packaging and release validation for <PRIVATE_REPO_D>. Use when building distributables, fixing packaging regressions, or validating installer/portable outputs without introducing CI workflows.
- ▌ Skill Creator Openclaw · grtninja bundleCreate or update AgentSkills. Use when designing, structuring, or packaging skills with scripts, references, and assets.
- ▌ Blender Vrm Visible Fit · grtninja bundleRun a checkpointed, in-foreground Blender workflow for VRM clothing fit and rig transfer. Use when dress/shoe fitting must be reviewed live before export and destructive body edits are not acceptable.
- ▌ Repo B MCP Comfy Bridge · grtninja bundleCanonical MCP adapter and Comfy bridge operations for <PRIVATE_REPO_B>. Use when enabling MCP, validating shim.comfy resources/tools, operating workflow/pipeline submissions, or running fail-closed Comfy diagnostics with optional AMUSE and CapCut contract checks.
- ▌ Request Loopback Resume · grtninja bundleResume previously requested work deterministically after interruptions. Use when context switches, AFK gaps, or multi-lane pauses require checkpointed state, explicit next actions, and fail-closed resume validation.
- ▌ Shockwave Dashboard Ops · grtninja bundleOperate and debug the Shockwave dashboard/router lane when shim-first model routing, startup contracts, tool-bridge behavior, and operator handoff must stay aligned. Use when recent dashboard work touches startup verification, bridge services, authority metadata, or router-facing docs.
- ▌ Skills Cross Repo Radar · grtninja bundleDetect recent cross-repo work and produce deterministic JSON evidence for skill curation and routing decisions. Use when workflows span multiple repos and you need a bounded view of recent commits, touched files, and policy/contract-sensitive changes.
- ▌ Local Comfy Wan Multigpu · grtninja bundleOperate local-only ComfyUI Wan image-to-video workflows with MultiGPU placement, quality-first source staging, long-clip validation, and Explorer-based export handoff. Use when Codex needs to turn private local stills into local video artifacts without browser-only UX, while avoiding destructive crops, stray console windows, and GPU contention with chat models.