jbaruch

Pick and wire a planner-driven Koog 1.0 agent — either LLM-based (the LLM picks the next action each turn, optionally with a critic loop) or GOAP (a classical planner searches a typed state space toward a goal). Pulls `ai.koog:agents-planner`, constructs the planner strategy, and wires it into `AIAgent(...)`. Use when the user asks to "use a planner", "let the agent plan", "use GOAP", "build a planning agent", names any of `Planners.llmBased`, `Planners.llmBasedWithCritic`, `Planners.goap`, `PlannerAIAgent`, `agents-planner`, or describes an open-ended task whose step sequence depends on runtime context.

Get typed structured output from a Koog 1.0 agent — pick between `nodeLLMRequestStructured` (graph DSL, schema-driven JSON) and `responseProcessor` (top-level on the agent factory, simpler shape). Defines the `@Serializable` output class and wires it into the strategy or the factory. Use when the user asks to "return JSON", "get a typed response", "use structured output", "return a data class from the agent", or shows a `data class` they want as the agent's output.

Snapshot a running Koog 1.0 agent's state at arbitrary points and restore later — distinct from the persistence checkpoint loop in `add-persistence`. Snapshots are caller-triggered; persistence is automatic and continuous. Use when the user asks to "snapshot the agent", "save state at this point", "restore from a snapshot", or names the `agents-features-snapshot` module.

Add checkpoint-and-resume to a Koog 1.0 agent. Two modes — `runFromCheckpoint` for replay-only use without installing a feature, and the full Persistence feature when you need rolling checkpoints, replay-with-modifications, or planner-agent durability across restarts. Use when the user asks to "make the agent resumable", "save progress", "checkpoint the agent", "restart from where it left off", or describes a long-running workflow that may be interrupted.

Author a custom graph strategy for a Koog 1.0 agent — pick the right node types, chain tool execution correctly, build edges with the infix vocabulary, and reach for subgraphs (`subgraphWithTask`, `subgraphWithVerification`) when steps deserve their own model, prompt, or tool subset while still sharing the agent's message history. Use when the user asks to "write a custom strategy", "build a graph for the agent", "author a strategy DSL", "add a verify-and-fix loop", "use subgraphs", or describes multi-step orchestration that won't fit inside `singleRunStrategy()`. Subgraphs are not for context isolation — for an independent agent that does not see the parent's history, use `Skill(skill: "add-tool")` Step 3 (sub-agent-as-tool).

Author a Koog 1.0 agent as a typed pipeline of domain-modeled subtasks — tools sliced by access pattern (read / write / communication) into separate ToolSets, inter-subtask handoffs as `@Serializable` `@LLMDescription`-annotated data classes (not text prompts), each subtask wired with `subgraphWithTask[In, Out]` using its own model and tool subset, self-correction loops via `subgraphWithVerification[T]` + `CriticResult[T]`. The integration pattern Koog's own banking demo uses. Use when the user asks to "model the agent as a pipeline", "build a multi-stage agent with typed handoffs", "give each stage its own tools", "build a verify-and-fix loop with typed data", or describes a workflow with distinct phases that should hand structured data to each other.

Connect a Koog 1.0 agent to an MCP (Model Context Protocol) server, using the primary 1.0 Streamable HTTP transport — or fall back to SSE / stdio when the remote server doesn't speak Streamable HTTP yet. Adds the agents-mcp dependency, builds a ToolRegistry from the MCP server's exposed tools, and merges it with the agent's existing tool registry. Use when the user asks to "connect to an MCP server", "use the GitHub MCP server", "add MCP tools to my agent", "wire Playwright MCP" or similar. Assumes a scaffolded Koog 1.0 project.

Add a new tool to an existing Koog 1.0 agent. Pick the right registration style (@Tool + ToolSet annotation, Tool[TArgs,TResult] subclass, or sub-agent-as-tool), define the args, and wire the tool into the agent's ToolRegistry. Use when the user asks to "add a tool to my agent", "expose something to the LLM", "let the agent call a function", or "wrap this agent as a tool for another agent". Assumes a scaffolded Koog 1.0 project — for new projects start with the scaffold-agent skill.

Move every app reference from an old Hubitat device onto a new one — via Settings → Swap Device where possible, a virtual-device bridge or parking slot where the swap list blocks it, and a guided manual re-select where neither works. Use when the user wants to swap, migrate, replace, or move a device's references to a different device, re-home a device to a different hub over Hub Mesh, or asks why a device does not appear in the Swap Device list.

Diagnose Hubitat Z-Wave and Zigbee network problems — ghost/failed nodes, packet errors, weak routes, dead or unjoined Zigbee devices, devices that stopped responding to commands, a broken hub-mesh link, an unhealthy mesh. Use when the user wants to check mesh health, find ghost nodes, debug a flaky/slow/dead Z-Wave or Zigbee device, or figure out why the radio network misbehaves.

Safely remove a Hubitat device — enumerate where it's used, warn about the blast radius before deleting, verify references cleared after, and re-wire them onto a replacement device. Use when the user wants to remove, delete, retire, or replace a device, or asks what a device is used by before deleting it.

Update Z-Wave device firmware on a Hubitat hub via the native zwaveJS updater — discover installed versions, find vendor-latest firmware, stage it, and batch-flash safely with a radio-hang watchdog. Use when the user wants to update/flash device firmware, check which devices are behind on firmware, or fix a device whose issue a firmware update addresses.

Tail a Hubitat hub's live log or event websocket, filtered, and interpret it against the code to diagnose an app or driver. Use when the user wants to debug, watch logs, tail the log stream, see live events, or figure out why a Hubitat app/driver misbehaves.

Deploy a Hubitat app or driver's Groovy source to a hub and confirm it saved and runs by watching the log stream. Use when the user wants to deploy, push, upload, or install app/driver code onto a Hubitat hub, or iterate the edit-deploy-check loop.

Manage the hubs.json config that records how to reach each Hubitat hub by IP for code operations. Actions — init a config, add a hub, set the default hub, remove a hub, list hubs. Use when the user wants to configure, register, add, or list Hubitat hubs for deploy/pull/debug.

Run the Hubitat sandbox linter on app or driver Groovy and judge each finding — real defect vs. false positive — before the code goes near a hub. Use when the user wants to lint, check, or validate Hubitat code, or automatically before deploying.

Create or update project governance principles and constitution

Create feature specification from natural language description

Identify underspecified areas and ask targeted clarification questions

Create technical implementation plan from feature specification

Generate domain-specific quality checklists for requirements validation

Generate test specifications from requirements before implementation (TDD support)

Generate actionable task breakdown from plan and specification

Validate cross-artifact consistency between spec, plan, and tasks

Promote agent-created skills and rules from NAS staging to tile GitHub repos via a full PR lifecycle — opens a PR, summons Copilot, iterates fixups until the review is clean, then merges so GHA publishes. Use when there are new items on staging, after check-staging shows pending items, or when asked to deploy skills, push to production, or publish rules to a tile repo.

Register a new Docker Compose project on UGOS Pro (NASync) when the compose file lives in the nanoclaw repo. Plumbs the `/volume1/docker/PROJECT_NAME` directory symlink, the in-repo `.env` symlink, and the UGOS Pro SQLite registration row so the project appears in the Projects UI without UGOS rewriting the tracked compose file. Use when adding a new sidecar that needs UGOS Pro UI Start/Stop visibility, when wiring a repo-tracked compose project onto the NASync for the first time, when migrating an existing service to the symlinked-compose topology, or when asked to "register a UGOS project" / "add a sidecar to UGOS Pro".

Sequential workflow for migrating an admin-tile skill, rule, or script set into a per-chat overlay tile. Audits cadence frontmatter, state-plane couplings, and cross-skill imports; moves files across two tile repos; updates per-group additionalTiles config; ships each side through publish-tile; verifies live materialisation. Use when extracting an admin skill to an overlay, refactoring admin content into per-chat tiles, splitting capabilities out of nanoclaw-admin, or wiring additionalTiles for a freshly extracted overlay.

PR-based lifecycle for shipping a code change through the NanoClaw fork chain. Covers the full path on private (jbaruch/nanoclaw) — create PR, summon Copilot, wait for review, fix CI + reasonable feedback, merge, clean up branches — then cherry-picks what qualifies to public (jbaruch/nanoclaw-public) and repeats the same lifecycle there. Enforces the scrub rules from repo-chain.md. Use when a code change is committed and needs to go out, when asked to ship a fix, open a PR, push to production, merge changes, or propagate a fix from private to public.

Sync private NanoClaw improvements to the public fork. Runs the scrubbed export script, creates a PR for review, and optionally merges. Use when private has accumulated fixes that should go public, after a batch of improvements, when explicitly asked to sync or export to public, or when asked to push changes or update the public repo with the latest private work.

Pull upstream updates into private NanoClaw. The chain is qwibitai → public → private. This skill handles both pulling qwibitai changes into public and then merging public into private. Use when upstream has new features, when the user asks to update, or when /update-nanoclaw is invoked.

List pending skills and rules on the NAS staging area. Shows what the agent has created or updated that hasn't been promoted to tiles yet. Use before running promote, or when the user asks what's on staging.

Kill a running agent container on the NAS by Telegram group JID. The orchestrator respawns a fresh container on the next message. Does NOT delete registration or group folder. Use when a container is stuck, stale, or needs a fresh start.

Create feature specification from natural language description

Identify underspecified areas and ask targeted clarification questions

Create technical implementation plan from feature specification

Generate domain-specific quality checklists for requirements validation

Generate test specifications from requirements before implementation (TDD support)

Generate actionable task breakdown from plan and specification

Validate cross-artifact consistency between spec, plan, and tasks

Execute implementation plan by processing all tasks in tasks.md

Fetch Trakt.tv watch history (shows, movies, ratings) for analysis and recommendation workflows. Use when the user asks for show recommendations, what to watch, wants to see their watch history, or wants suggestions based on their viewing habits.

Weekly entertainment refresh: pulls Trakt watch history, checks watchlist for releases, runs show + book recs, syncs Audible purchases. Triggers: 'entertainment sync', 'weekly entertainment', 'sync trakt audible recs', 'refresh entertainment'.

Analyzes Baruch's viewing history and explicit ratings across netflix-history.csv, imdb-ratings.csv, and trakt-history.json to identify preferred genres, classify completed and abandoned shows, and rank unwatched titles by predicted interest. Generates targeted TV show recommendations with quality thresholds, searches for new releases, and tracks upcoming shows in watchlist.json. Use when Baruch asks for show recommendations, "что посмотреть", "что смотреть", or similar requests for what to watch next.

Back up new Audible audiobook purchases, decrypt to M4B, and append to books-library.csv. Use on "audible backup", "check new audiobooks", "sync audible library", or from weekly scheduled task.

Weekly fetch of recent comments on Baruch's YouTube channel; sends a per-video summary if new comments appear, silent otherwise. Triggers: 'check youtube comments', 'youtube comment check', 'fetch youtube comments', 'review channel comments'.

Checks tracked upcoming TV shows in watchlist.json and sends a Telegram notification message via MCP when any have been released. Use when running nightly release checks, monitoring streaming release dates, or checking whether new episodes or shows from a watchlist are now available to watch. Fires nightly via its own scheduled_tasks row (post-#404).

Recommend audiobooks based on Baruch's reading history and preferences. Analyzes reading patterns, filters unread titles by genre and rating, identifies series continuations, checks for new releases from favorite authors, and suggests similar authors or highly-rated unread titles from the Audible library JSON. Use when Baruch asks for book recommendations, "what to read next", wants something similar to a specific author, or asks about his unread queue.

Parses presentation talks to catalog rhetoric patterns: opening hooks, humor style, pacing, transitions, audience interaction, slide design, and verbal signatures. Downloads YouTube transcripts and analyzes slides (from PPTX, Google Drive PDFs, or video extraction), examining HOW the speaker presents. Processes talks in parallel batches and updates the running rhetoric summary. Triggers: "parse my talks", "run the rhetoric analyzer", "analyze my presentation style", "how many talks have been processed", "update the rhetoric knowledge base", "check rhetoric vault status", "process remaining talks for style patterns".

Creates new presentations grounded in the speaker's documented rhetoric patterns, using a personal rhetoric-knowledge-vault as a constitutional style guide. Follows an interactive, spec-driven process: distill intent from the user's prompt, jointly select rhetorical instruments from the vault catalog, architect the talk structure, develop content with speaker notes, run guardrail checks, generate a .pptx deck, and publish per the speaker's workflow. Use this skill whenever the user wants to create a new presentation, build a talk, write a conference submission, design a slide deck, prepare for a speaking engagement, or mentions "presentation" or "talk" in the context of content creation. Also trigger when the user describes a topic they want to present on, asks to adapt an existing talk for a new audience, or wants to develop a CFP abstract. Not a generic slide-deck tool — requires a populated rhetoric-knowledge-vault and follows the speaker's established style.

Generates or updates the structured speaker-profile.json from vault data. Aggregates rhetoric summary, slide design spec, confirmed intents, and structured talk data into a machine-readable profile used by the presentation-creator skill. Also generates speaker achievement badges. Triggers: "generate speaker profile", "update speaker profile", "regenerate speaker profile", "sync speaker profile".

Generates the visual layer of a talk: deck illustrations (FULL / IMG+TXT slides with a shared style anchor), progressive-reveal build chains, and YouTube thumbnails. Owns style-strategy collaboration (informed by the speaker's visual_style_history in the vault), prompt safety, edit-vs-regenerate asymmetry, build chaining, title-safe-zone composition, and thumbnail composition with a real speaker photo. Invoked by presentation-creator during illustration strategy (Phase 2), illustration generation and application to the deck (Phase 5), and the post-event YouTube thumbnail (Phase 7). Triggers: "illustrate the deck", "generate illustrations", "create slide visuals", "design the visual style", "make a thumbnail", "build a YouTube thumbnail", "add visuals to my talk", "regenerate slide image", "fix the thumbnail", "generate progressive reveals", "build sequence for a slide".

Publish a talk page to the Jekyll-based shownotes site (e.g., speaking.jbaru.ch). Composes a markdown file in the site's `_talks/` collection so the custom Jekyll parser plugin extracts the right fields, the talk.html layout renders correctly, and the "Video Coming Soon" badge fires when the recording isn't ready yet. Use when the user says "publish shownotes", "create shownotes page", "add talk to shownotes", "shownotes for [some talk]", "shownotes site", "speaking.jbaru.ch", or asks to update a talk page (e.g., "add the video to the shownotes", "the video is out", "update shownotes with the recording"). Also trigger for first-time publishing before a talk is delivered, when only the slides URL exists. The Jekyll site at `~/Projects/shownotes` uses a custom markdown parser (`_plugins/markdown_parser.rb`) that imposes specific format rules — this skill encodes those rules so the agent doesn't author content that silently fails to render.

Runs interactive clarification sessions with the speaker after talk processing. Resolves ambiguities in rhetoric observations, validates findings, captures speaker intent, conducts humor post-mortems, and probes for blind-spot moments invisible to transcripts. Stores confirmed intents and infrastructure config in the tracking database. Triggers: "run clarification session", "humor post-mortem", "blind spot review", "capture speaker intent", "clarify rhetoric findings".

Travel-data refresh bundle: TripIt → Reclaim timezone sync, refresh travel-schedule.json from the TripIt iCal feed with a two-tier Gmail freshness probe, rebuild travel-db.json, log newly-appeared trips to daily memory, then check upcoming trips for booking gaps. Runs daily; precheck-gated on travel-db.json freshness. Triggers: 'sync trips', 'sync travel', 'update travel data', 'pull trip info', 'refresh travel schedule', 'rebuild travel db', 'check my bookings'.

On a byAir precheck wake event, reconciles the operator's managed calendar events (boarding block, adopted byAir flight event, Reclaim travel-block cleanup, switched-away teardown) and composes a user-facing flight notification — delay, gate change, cancellation, boarding, connection risk, inbound-delay, time-to-leave, baggage carousel, day-before check, or arrival logistics. Also configures the plugin (verify credentials, set home base). Use when a tracked-flight wake event needs a notification, or when setting up or diagnosing flight-assist. Triggers - "check flight-assist env", "diagnose flight-assist", "set flight-assist home base", "set home address", "configure flight-assist", "flight delay notification", "gate change notification", "cancellation notification", "boarding alert", "time to leave alert", "inbound delay notification", "baggage carousel", "arrival logistics", "day before sanity check", "flight removed upstream", "connection at risk", "tight connection alert", "reconcile calendar".

Shared travel-domain library bundle for the nanoclaw-travel plugin. Hosts the cross-skill Python modules trip_origin (TripIt-over-home position/anchor resolution) and airport_lead (clearance / post-arrival buffer policy) so flight-assist and the drive engine import one source of truth instead of duplicating them. Not a user workflow — background code the other skills load at runtime; never invoke directly.

The unified drive-block engine: manages the travel-time / driving blocks on your primary calendar from one place — the drives to your flights and the drives to your in-person meetings. On a schedule it plans both, diffs them against the blocks already there, and applies the changes — adding, updating, and removing its own blocks — suppressing drives you can't make (an airport reached by a connection, a home meeting while you're travelling). Use when the user asks about drive blocks, driving time, commute or travel-time blocks on their calendar, or drives to the airport or to a meeting; when the operator replies to a drive notification to skip a meeting drive they are not making ('skip', 'skip 1', 'skip 2 and 3', 'skip the Massage drive'); also runs on its own schedule and wakes you only when it added a skippable meeting drive or a drive time materially changed.

Checks upcoming trips for missing bookings (flights, hotels, accommodation) by reading the nightly-built `travel-db.json`. Reports gaps for all upcoming trips — no date limit. Supports snooze state. Silent when all bookings are complete or snoozed. Use when the user asks about upcoming travel plans, itinerary completeness, missing reservations, or TripIt trip status.

Adaptive scheduler for the TripIt/byAir refresh of active-flights.json. Precheck-gated to keep byAir polling responsive on flight days and idle between travel windows. Use when active-flights.json isn't updating, byAir polling cadence isn't matching flight density, troubleshooting flight tracking / flight notifications / flight status updates / travel schedule refresh, or setting up flight-assist on a new install. The gate predicate and threshold constants live in precheck.py.

Run the A1 pipeline (produce-mode discovery → auto-selection → build-and-evaluate) across a CSV of company names. Drives the full pipeline unattended for booth-scale runs (~100 leads), sharing a single DISCOVERY_RUN_TS so all per-company outputs land in one diffable run directory. After completion, emits a per-batch `index.md` summarizing outcomes (BUILD/SKIP/AMBIGUOUS counts, aggregate lift across BUILDs, links to per-company reports). Trigger phrases — "run the batch driver on a CSV", "process the attendee list", "batch-pipeline the snyk leads", "A1 batch run on a CSV of company names".

Steps 4–10 of the workflow — takes a selection.json produced by `select-target` and runs the full skill-build + eval + report pipeline. Scaffolds a skill via `tessl skill new`, authors its body for the selected target, generates eval scenarios via `tessl scenario generate`, audits them for bleeding/leaking, runs `tessl skill review --threshold 85` as a quality gate, runs `tessl eval run --variant without-context --variant with-context` to get baseline + with-skill scores in one call, parses the result, computes per-scenario lift, analyzes gaps, and writes a markdown report alongside the discovery / selection artifacts. End-to-end output for the A1 MVP cell of the operating-modes 2x2 in `SPEC.md`. Trigger phrases — "build the skill", "run phases 4-10", "build-and-evaluate", "generate and evaluate the skill for a given company slug".

Triage a raw list of company names (conference attendees, prospect lists, sales lists) before running the discovery pipeline. Classifies each company into seven buckets and returns a focused candidate list of companies whose discovery is likely to yield a usable skill target. Use whenever the user provides a list of more than a handful of company names and needs to decide which deserve discovery effort. Trigger phrases — "filter this list", "triage these companies", "pre-discovery filter", "narrow the conference list".

Produce-mode automated source discovery for a single company (steps 1–2 of the workflow, A1 cell of the operating-modes 2x2 in `SPEC.md`). Takes a company name (optionally `parent/sub-brand` for MEGA_CORP cases that have been pre-scoped) and produces a discovery JSON conforming to `discovery-output-contract.md` — schema_version 3, mode=produce. Targets the company external/public API/SDK surface — what an outside developer would consume — rather than the internal dogfood surfaces consume-mode targets. Output is the input to step 3 (human-gated target selection). Trigger phrases: "run produce-mode discovery on X", "discover external API surface for X", "A1 discovery on X", "produce a produce-mode discovery JSON for X".

Automated source discovery and structure extraction for a single company (steps 1–2 of the workflow in `SPEC.md`). Takes a company name (optionally `parent/sub-brand` for MEGA_CORP cases that have been pre-scoped), produces a JSON artifact conforming to `discovery-output-contract.md`, and writes it to a versioned per-run directory. Output is the input to step 3 (human-gated target selection). Trigger phrases — "run discovery on X", "discover sources for X", "produce a discovery JSON for X".

Step 3 of the workflow — human-gated target selection from a discovery.json. Reads a discovery output, presents ranked skill_targets to the human (sorted by booth-aha score for consume-mode v2 and v3+consume, by raw confidence for v1 and produce-mode v3), accepts a single pick, and persists a selection.json alongside the discovery file. The only manual step in the pipeline per `SPEC.md`. Trigger phrases — "pick a target", "select skill target", "human-gate selection", "step 3 selection", "run selection on a company slug".

Build a log-style or chat-style scrollable pane in a TamboUI Toolkit app — `ListElement` configured with no selection highlight, sticky scroll, scrollbar, mouse-wheel capture, focus chain integration, and a pre-wrap helper for long content. Use when the user says "build a log pane", "add a chat pane", "make a scrollable list", "trace pane that auto-scrolls", "tail-style output", or asks how to display many lines of streamed text in a TUI.

Wire keyboard and mouse focus across multiple panes in a TamboUI Toolkit app — assign stable ids, set the initial focus in `onStart()`, and add a visible focused-pane border via `focusManager.focusedId()`. Use when the user says "add multi-pane focus", "set initial focus on the input", "highlight the focused pane", "tab between panes", "make the prompt focused by default", or asks how to manage focus across panels.

Add a new Java Flight Recorder event to a TamboUI module following project conventions — `dev.tamboui.AREA.THING` naming, `enabled()` guards, static `commit(...)` helper, and `compileOnly(libs.jfr.polyfill)` for Java 8 modules. Use when the user says "add a JFR event", "trace X with JFR", "instrument Y for flight recorder", or "emit a JFR event for Z".

Bootstrap a new TamboUI Toolkit DSL application — generates a `ToolkitApp` subclass with a working `render()` and a `main` entry point, plus the right Maven/Gradle/JBang dependencies. Use when the user says "create a TamboUI app", "scaffold a TUI", "new tamboui app", "hello world TamboUI", or asks to start a TUI project from scratch.

Add a Toolkit `Element` wrapping an existing TamboUI widget so it gains CSS support, styled sub-components, focus integration, and a Toolkit factory method. Use when the user says "wrap a widget", "add an element for X widget", "make widget Y CSS-aware", "expose widget Z in the DSL", or asks to integrate a custom widget into the Toolkit.

Native Google Calendar and Tasks reads over the OneCLI gateway, for trusted-tier ground-truth verification. Run google-calendar.py events-list to read the owner's calendar, or google-tasks.py list-tasklists / list / get to read task lists and task status. The OneCLI gateway injects the Bearer on the wire — no Google credential in the container. Gmail is intentionally excluded. Use when a trusted agent must verify a calendar event or a task/todo status, or otherwise read the owner's Calendar or Tasks.

Session bootstrap and rolling memory updates for trusted containers. On session start, reads MEMORY.md (permanent facts), RUNBOOK.md (operational workflows), recent daily and weekly logs, and highlights.md to restore context. After non-trivial interactions, appends timestamped entries to group-local and cross-group shared daily logs. Use when starting a new session to load previous notes and remember context, or after meaningful conversations to save conversation history, persist session state, or record newly learned owner preferences.

Read-only system-status probe for trusted-tier NanoClaw containers — surfaces stuck scheduled tasks, DB size, and recent task-run failures from the orchestrator's SQLite at `/workspace/store/messages.db`. Use as part of heartbeat or standalone. Triggers on "system status", "check tasks", "stuck tasks", "database size", "task failures".

Quick read-only health check — session context, workspace mounts, tool availability, and task snapshot. Use when the user asks for system status, health check, diagnostics, system info, check environment, what tools are available, or runs /status.

Migrate a legacy tile.json plugin to the .tessl-plugin/plugin.json form. Runs `tessl plugin migrate`, renames .tileignore to .tesslignore, removes the obsolete tile.json, re-lints, then reconciles residual "tile" wording to "plugin" in the repo's prose. Use when a repo still has a tile.json (and no .tessl-plugin/plugin.json), or when the user asks to migrate / convert / modernize a tile.json plugin, move off tile.json, or adopt the plugin.json manifest form.

Structured workflow for shipping code via GitHub pull requests: PR creation, dual-lens automated review (gh-aw for `rules/*.md` compliance + Copilot for doc accuracy and cross-step consistency), merge, and branch cleanup. Covers readiness checks, version reasoning, review polling, feedback handling, and post-merge verification. Use when the user wants to open a pull request, ship code, merge a branch, or handle post-merge cleanup on GitHub.

Scaffold the `jbaruch/coding-policy` gh-aw PR review workflows into a consumer repository: copies the packaged paired workflow templates (OpenAI + Anthropic reviewers), compiles them with `gh aw`, and opens a PR. After merge, every pull request in the repo is reviewed against the latest published `jbaruch/coding-policy` rules by the reviewer whose family differs from the PR's declared author model — avoiding self-review bias per `rules/author-model-declaration.md`. Use when the user wants to add, install, enable, scaffold, set up, or wire up an automated policy review / PR reviewer / coding-policy CI reviewer in a consumer repo. Also use to upgrade, update, refresh, or pull the latest reviewer templates into a consumer repo that already has the scaffold installed — the skill switches to override mode in that case.

Use when the user asks to check, review, look at, or act on a pull request by number — "check PR 5", "review PR 12", "what's going on with PR 7", "look at this PR". Classifies the PR as same-repo or fork. Fork PRs are skipped by the gh-aw policy reviewer's fork-guard; this skill offers to adopt the fork branch into the base repo as a same-repo PR the reviewer can run on, preserving the contributor's commits. Same-repo PRs pass straight through to the existing reviewer.