skill-arbiter

Run the local NullClaw host security app for skill governance, curated-source discovery, guarded threat suppression, and self-governance on Windows hosts.

Sweep one or more local repositories for implementation gaps and produce deterministic, evidence-backed remediation lanes. Use when you need cross-repo detection of missing tests, docs lockstep drift, risky TODO/FIXME additions, and release-hygiene misses.

Build and query bounded metadata-only repository indexes without rg process churn. Use when working in very large repos and you need deterministic file discovery by path, extension, language, scope, or freshness.

Detect 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.

Enforce hardware-first diagnosis and fixes in <PRIVATE_REPO_B>. Use for runtime probe, telemetry, inference, and integration failures where strict real-hardware behavior, no new stubs, deterministic diagnostics, and no unrequested driver/runtime mutation are required.

Run a defender-first security sweep on code, configs, prompts, model/tooling surfaces, or third-party contribution lanes. Use when a request involves safe bug, leak, zero-day-class, exploit, or hack hunting for protection, when contributing to outside repositories and you want a focused security pass, or when touching auth, secrets, permissions, network exposure, prompt/tool boundaries, data flow, or update/build surfaces. This skill is defensive only and must never be used for weaponization or unauthorized access.

Present, navigate, snapshot, and hide HTML content on connected OpenClaw nodes (Mac, iOS, Android) via the canvas tool. Use when displaying games, visualizations, dashboards, or interactive demos on remote node canvas views, or when debugging canvas URL routing across loopback, LAN, and Tailscale bind modes.

Keep 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.

Read, write, append, and manage Feishu/Lark documents via the feishu_doc tool. Use when creating, reading, or editing Feishu Docx documents, managing document blocks, creating tables, uploading images or file attachments, or extracting doc_token from Feishu URLs.

Browse and manage Feishu Drive folders and files via the feishu_drive tool. Use when listing shared folders, inspecting file metadata, creating folders, moving files, or deleting Drive items the bot can access.

Manage Feishu file and document permissions via the feishu_perm tool. Use when listing collaborators, granting access, revoking access, or checking whether a file, folder, wiki node, or document has the correct permission level.

Navigate and manage Feishu Wiki spaces and nodes via the feishu_wiki tool. Use when listing spaces, browsing wiki trees, creating or moving wiki pages, renaming nodes, or resolving a wiki node into the backing Docx object for feishu_doc edits.

Execute multi-step workflows with approval checkpoints via the Lobster pipeline engine. Use when the user wants repeatable automations (email triage, PR monitoring, scheduled syncs), when actions need human approval before executing (send, post, delete), or when multiple tool calls should run as one deterministic pipeline operation.

Apply 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.

Align 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.

Align desktop autonomy, speech fallback, live research, and chat-route runtime behavior across avatar-facing local apps after model-plane or harness shifts. Use when runtime handoff and UI behavior change together.

Operate AvatarCore bridge sessions, heartbeats, engine-event queues, and directive drains as one contract. Use when Unreal bridge lifecycle and queue semantics change together.

Enforce AvatarCore workstation desktop-launch acceptance when frontend/backend launcher ownership, stale-window cleanup, and health proof must be explicit. Use when backend-only startup is being mistaken for app readiness or when a combined launcher is being introduced or repaired.

Govern queue ownership, backpressure, retry and cancel semantics, and mutual-exclusion behavior across STARFRAME local apps. Use when `9040`, `9041`, bridge queues, or operator queue surfaces change together.

Review recent open diffs across multiple repos, classify repeated release/docs/tests gaps, and route each gap to the right remediation skill. Use when several repos are carrying unreleased work and you need a deterministic audit before patching or curating new skills.

Audit 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.

Keep local training dataset provenance, merged outputs, and manifest truth aligned across dataset-building repos. Use when teacher distillation, enrichment exports, or study manifests change together.

Govern the local-first Penny speech plane across ASR, TTS, endpoint voice behavior, and backend action routing. Use when Shockwave, AvatarCore, VRM Sandbox, or control-plane voice contracts shift together.

Admit 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.

Fetch GitHub issues, spawn sub-agents to implement fixes and open PRs, then monitor and address PR review comments. Usage: /gh-issues [owner/repo] [--label bug] [--limit 5] [--milestone v1.0] [--assignee @me] [--fork user/repo] [--watch] [--interval 5] [--reviews-only] [--cron] [--dry-run] [--model glm-5] [--notify-channel -1002381931352]

Validate 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.

Enforce local-first Codex execution through VS Code workspace and connected local apps/services/hardware, with MemryX shim priority and fail-closed remote-host checks.

Correlate 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.

Operate 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.

Govern Media Workbench indexing posture across metadata-only baselines, opt-in content-derived indexing, local manifest storage, and operator-facing index summaries. Use when indexing tools, manifests, catalog DBs, or media-root governance change together.

Guard the STARFRAME Media Workbench worker ownership contract, queue/work-item lifecycle, and desktop-to-worker status surfaces. Use when changes touch `9040`, job payloads, resumable work items, operator snapshots, or worker-facing docs/tests.

Orchestrate app-first, registry-aware bring-up and health validation for local multi-app meta-harness stacks. Use when startup order, runtime registry, launcher scripts, control-center bring-up, or cross-app health checks change together.

Align 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.

Split 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.

Operate 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.

Build, enrich, distill, and validate local training datasets for the Qwen Training Workbench. Use when dataset export, weighted-study assembly, label distillation, workstation-mode enforcement, or training-input validation change together.

Operate 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.

Run credit-first local Agent Bridge orchestration in <PRIVATE_REPO_B> with strict read-only validation, bounded indexing, and fail-closed guidance hints. Excludes MCP Comfy diagnostics.

Operate 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.

Enforce one-brain routing across VRM Sandbox, PC Control, and MemryX shim with shim-first public endpoint authority on port 9000. Use when VRM chat/control behavior must be governed by PC Control without changing the public model-plane endpoint.