kingkillery

Coordinate split-view main-agent and sub-agent workflows using a shared file-based command queue. Works with ANY CLI IDE that can read/write files and invoke skills. One instance assumes the main-agent role (coordinator); N other instances take on sub-agent roles (workers). Use when the user wants to run a side sub-agent in parallel to the main agent, dispatch research or exploration tasks to sub-agents, poll for sub-agent results, or manage multi-agent split-view sessions. Trigger on "run side agent", "launch sub-agent", "main-agent mode", "sub-agent mode", "split view workflow", "coordinate sub-agent", or when setting up agent-to-agent task delegation.

This skill should be used when the user asks to "compare candidate patches", "rank multiple solutions", "choose between implementations", "verify two drafts", "pick the best trajectory", "use llm-as-a-verifier", or wants repeated criteria-decomposed LLM verification instead of a single free-form judgment.

Run the QuickScope stage of qs-parallelprd: turn a messy objective, findings list, repo issue, or research-backed gap analysis into an evidence-rooted A-to-Z lane plan with root cause, lane ownership, dependencies, parallel windows, serialization rules, effort, and targeted acceptance gates.

A relentless interview to sharpen a plan or design.

Interview the user relentlessly about a plan or design. Use when the user wants to stress-test a plan before building, or uses any 'grill' trigger phrases.

Compact the current conversation into a handoff document for another agent to pick up.

Teach the user a new skill or concept, within this workspace.

Reference for writing and editing skills well — the vocabulary and principles that make a skill predictable.

Voice-driven code agent. Talk to your codebase in natural language with speech-friendly responses. Triggers on "codechat", "voice chat about code", "talk to my codebase", or "chat about code".

Use when the user wants the KADE plus gstack bridge, including session kickoff with user-profile context, /g-kade install style project scaffolding, repo adaptation, or structured handoff logging tied to llm-wiki-memory packet setup. This packet-owned wrapper is installable into ~/.agents/skills, ~/.codex/skills, and ~/.claude/skills.

Use when the user wants gstack-style execution workflows such as QA, browser dogfooding, code review, debugging, ship or PR preparation, design review, benchmark checks, or deployment verification. This packet-owned wrapper is installable into ~/.agents/skills, ~/.codex/skills, and ~/.claude/skills.

Create distinctive, production-grade interfaces with high design quality using the Impeccable frontend vocabulary. Use when designing polished web or Android app experiences, especially when the user wants stronger visual taste, clearer UX, or a more memorable aesthetic.

Use when the user wants the KADE System itself, including Layer 1 and Layer 2 context loading, KADE session structure, HUMAN.md and kade/ overlays, or project initialization and handoff workflows. This packet-owned wrapper is installable into ~/.agents/skills, ~/.codex/skills, and ~/.claude/skills.

Sync agent skills from a canonical source directory into one or more Obsidian vaults or project workspaces. Compares skill directories by file count and total size, reports new/updated/vault-only skills, then copies everything from canonical into the target while preserving target-only skills. Triggers on "sync skills", "update skills", "skill sync", "refresh skills", "pull latest skills", "sync vault skills", "check for skill updates", or "are my skills up to date".

Expert guide for the NotebookLM CLI (`nlm`) and MCP server - interfaces for Google NotebookLM. Use this skill when users want to interact with NotebookLM programmatically, including: creating/managing notebooks, adding sources (URLs, YouTube, text, Google Drive), generating content (podcasts, reports, quizzes, flashcards, mind maps, slides, infographics, videos, data tables), conducting research, chatting with sources, or automating NotebookLM workflows. Triggers on mentions of "nlm", "notebooklm", "notebook lm", "podcast generation", "audio overview", or any NotebookLM-related automation task.

Voice extension for Pi with spoken replies, wake-word listening, remote control, and Telegram integration. Use when user mentions /speak, /mono, /phone, /remote, voice output, spoken replies, or pi-speak extension. Install: pi npm i pi-speak-pk

Pointer skill for pk-speak / pi-pk-speak. Use when the user wants spoken replies, wake-word listening, session-manager control, deterministic short voice routes like `PK one` / `PK1` vs `PK two` / `PK2`, Telegram control, QR-based Android setup via `/pk-remote`, or the browser remote. Start from the bridge doc and then jump to README and source as needed.

How agents execute work for Kade. Use when starting substantive or multi-step work, discovering workspace and repository instructions, selecting skills or tools, assembling minimal context, delegating or handing off work, verifying results, or synchronizing state. Do not use for simple direct answers, Kade's profile or governance (use kade-hq), or repository-specific implementation truth (use repo-local instructions).

Canonical human-context and governance skill for working with Kade (pk). Selects relevant communication, attention, autonomy, session-state, decision, commitment, and memory rules without forcing planning behavior onto ordinary execution.

Survey any codebase as a senior advisor and produce prioritized, self-contained implementation plans for OTHER models/agents to execute. Strictly read-only on source code — never implements, fixes, or refactors anything itself. Use when asked to audit a codebase, find improvement opportunities (bugs, security, performance, test coverage, tech debt, migrations, DX), suggest features or where to take the project next (roadmap, product direction), or generate handoff plans for another agent to implement.

Use when a CLI agent should speak naturally with pk-speak, wrap a command with spoken lifecycle notices, start the phone gateway/tray, print Android setup QR codes, or explain the current realtime voice options. Keep normal replies concise and speak only useful status, questions, approvals, and completion summaries.

Use the multimodalart/higgs-audio-v3-tts Gradio Space via API. Provides Python, JavaScript, and cURL usage examples.

Use when the task should consult, capture, validate, evolve, or retire reusable workflow shortcuts through the local `llm-wiki-skills` MCP server or its CLI. This packet-owned wrapper explains the full tool lifecycle in depth and is installable into ~/.agents/skills, ~/.codex/skills, ~/.claude/skills, and ~/.pi/agent/skills.

Repo-local gstack workflow surface layered on top of the llm-wiki-memory packet.

Use when the user wants the KADE System itself, including Layer 1 and Layer 2 context loading, KADE session structure, HUMAN.md and kade/ overlays, or project initialization and handoff workflows. This packet-owned wrapper is installable into ~/.agents/skills, ~/.codex/skills, and ~/.claude/skills.

Use when the task should consult, capture, validate, evolve, or retire reusable workflow shortcuts through the local `llm-wiki-skills` skill/CLI surface. MCP is optional transport only. This packet-owned wrapper explains the full tool lifecycle in depth and is installable into ~/.agents/skills, ~/.codex/skills, ~/.claude/skills, and ~/.pi/agent/skills.

Repo-local harness for evaluating agents, classifying failures, validating regressions, and producing review artifacts.

Use this skill when the repo is an llm-wiki-memory vault and the task is to ingest a source, answer from the persistent wiki, create or review reusable skills, use pk-qmd for repo-local evidence retrieval, consult brv for durable memory, or lint and maintain the wiki.

Use the repo-local harness to benchmark Gym-Anything-compatible agents, cluster failures, and generate human-gated improvement proposals.

Run the repo-local harness to evaluate agents, classify failures, propose human-gated improvements, validate regressions, and generate review packets.

Use when the user wants the KADE plus gstack bridge, including session kickoff with user-profile context, /g-kade install style project scaffolding, repo adaptation, or structured handoff logging tied to llm-wiki-memory packet setup. This packet-owned wrapper is installable into ~/.agents/skills, ~/.codex/skills, and ~/.claude/skills.

Use when the agent should bootstrap a neutral repo with llm_wiki_prompt_packet and launch the Gym-Anything Pokemon benchmark through the packet-owned wrapper surface. This skill handles the contract-first smoke run, the optional framework run, and the required machine-readable result reporting path.

Run one Codex-managed autopilot iteration using Ruflo MCP or CLI state

Use learned patterns and current state to predict the optimal next Codex action

Discover and recommend ruflo plugins based on your workflow, installed MCP tools, and current task

Initialize a new Ruflo project with MCP tools, hooks, and agent configuration

Observe Ruflo swarm status from Codex with bounded checks

Initialize a Ruflo swarm and coordinate Codex-native follow-up work

Create a new sequentially numbered database migration with up/down SQL files

Generate and maintain documentation with drift detection

Red-team reviewer that tries to break code and assumptions via adversarial inputs.

Workflow skill for the PumpFun Bundler watcher/GUI system. Use when working on gui.ts, the watcher supervisor, watcher subprocess (task43-watch-pump-program.ts), or the /api/watch/* routes. Covers: starting/stopping the GUI, watcher lifecycle, state file IPC, route patterns, environment variables, and safe commit workflow.

Code review practices with technical rigor and verification gates. Use for receiving feedback, requesting code-reviewer subagent reviews, or preventing false completion claims in pull requests.

Systematic workflow for planning and implementing large features in phases with Spec Mode, validation, and clean sessions.

Analyze codebase with parallel Explore agents to produce .planning/codebase/ documents. Use for brownfield project onboarding, refreshing codebase understanding after significant changes, before major refactoring, when onboarding to unfamiliar codebases, or when asked to understand codebase or analyze project structure. Creates structured documentation of stack, architecture, structure, conventions, testing, integrations, and concerns.

Obsidian plugin & theme development reference. Read only the specific reference files you need for the current task.

Author, review, and optimize notes and structure using Obsidian-flavored Markdown (OFM) and vault-safe conventions. WHEN: User asks for vault optimization, note refactors, OFM syntax, linking/backlinks, properties, or note organization. WHEN NOT: Generic markdown only (no Obsidian features), or non-documentation coding tasks.

Expands current-progress to final-product gaps into a multi-pass plan and closes those gaps with a fresh-context sub-agent. Use for quick-scope planning, 3-phase expansion, A to Z gap tracking, or fill-the-gap runs.

Adaptive certificate-driven reasoning for non-trivial engineering tasks in this repo, including semantic PR review, patch equivalence checks, debugging, safe implementation, refactors, codebase QA, and static risk review. Use when the task is cross-file, high-risk, behaviorally sensitive, or execution-constrained. Do not use for trivial mechanical edits or obvious single-file changes.

Adaptive certificate-driven reasoning for non-trivial engineering tasks in this repo, including semantic PR review, patch equivalence checks, debugging, safe implementation, refactors, codebase QA, and static risk review. Use when the task is cross-file, high-risk, behaviorally sensitive, or execution-constrained. Do not use for trivial mechanical edits or obvious single-file changes.

Adaptive certificate-driven reasoning for non-trivial engineering work in this repo, including semantic review, patch equivalence checks, debugging, safe implementation, refactors, test design, codebase QA, and static risk review. Use when the task is cross-file, behaviorally sensitive, risky, execution-constrained, or needs an auditable reasoning trail about what changed and why. Also use when improving repo-local agent instructions or the ACA skill itself. Do not use for trivial mechanical edits or obvious single-file changes.

Prepare and execute a Plannotator release — draft release notes with full contributor credit, bump versions across all package files, build in dependency order, and kick off the tag-driven release pipeline. Use this skill whenever the user mentions preparing a release, bumping versions, writing release notes, tagging a release, or publishing. Also trigger when the user says things like "let's ship", "prep a release", "what's changed since last release", or "time to cut a new version".

Analyze a user's Plannotator plan archive to extract denial patterns, feedback taxonomy, evolution over time, and actionable prompt improvements — then produce a polished HTML dashboard report. Falls back to Claude Code ExitPlanMode denial reasons when Plannotator data is unavailable.

This skill should be used when the user asks to "review plugin gaps", "audit each folder and file", "do a Clearwing-style gap analysis", "compare plugin structure maturity", or wants a per-folder and per-file structural review across one or more Obsidian plugin directories.

Set up Symphony (OpenAI's Codex orchestrator) for a user's repo. Use when the user mentions Symphony setup, configuring Symphony, getting Symphony running, or wants to connect their repo to Linear for autonomous Codex agents. Also use when the user says "set up symphony", "configure symphony for my repo", or references WORKFLOW.md configuration.

KADE-structured session orchestrator for gstack. Combines KADE's ADHD-friendly session structure, handoff logging, and user-profile context with gstack's execution skills (/ship, /review, /qa, /investigate, etc.). Invoke when starting a work session that needs both personal context-awareness and gstack task power. Also handles /g-kade install to scaffold KADE files into any project. (gstack)

Launch and inspect tf-grpo-upstream practice or eval workflows. Use when the task is to run an experiment, validate config wiring, check required environment variables, or summarize outputs from runs/<exp_id>.

Use this skill when configuring, updating, or operating Consurg for agent context control (scopes, permission tiers, wire integrations, tracing, and enforcement) in CLI workflows.

This skill should be used when the user asks to "create a scope", "restrict agent access", "set up consurg", "scope files", "add files to working set", "wire consurg to Claude", "wire consurg to Codex", "wire consurg to Gemini", "start the guard", "trace dependencies", "build scope from git diff", "export scope", "narrow scope for subagent", "check scope status", "limit what Claude can see", "restrict file access", "set up file permissions for agent", "protect files from agent edits", "block files from agent", "promote a file", "unblock a file", "show what files the agent can access", "scope my project", "export scope for cursor", "export scope for aider", "run with scope enforcement", "consurg", "pin my scope", or needs to manage file-level access control for AI coding agents using Context Surgeon's 5-tier permission model.

Project-scoped skill for keeping this repo faithful to the SpreadsheetLLM paper (arXiv:2407.09025). Triggers on requests to "check fidelity to the paper", "remediate gap analysis", "verify paper alignment", "rerun the eval pipeline", or to add a new gap-remediation iteration.