wcygan

Answer questions about how the pi coding agent itself works by reading pi's bundled offline docs and citing the file. Use when asked how pi does X, how to configure pi (settings.json, models.json, providers, keybindings), what a pi flag/mode/feature does, or where a pi feature is documented. Keywords pi docs, how does pi, configure pi, pi settings, models.json, pi providers, pi extensions, pi sessions, pi modes, pi skills.

Author Agent-Skills-standard SKILL.md files designed to be executed by the pi coding agent — often a small local model (llama.cpp/Qwen). Covers pi skill anatomy and discovery, writing low-freedom procedures with machine-checkable validators for small executors, the author-pilot-refine distillation loop driving pi headlessly, and optional Codex co-authoring. Use when writing or refining a skill that pi will run, or when asked to make a skill work well on a local model. Keywords pi skill, SKILL.md, pi coding agent, local model, llama.cpp, qwen, skill distillation, /skill, pilot, refine.

Use the pi coding agent CLI (@earendil-works/pi-coding-agent) and navigate its docs and source. Covers interactive/print/json/rpc modes, model & tool flags, sessions, custom models.json (local llama.cpp/Ollama/vLLM), skills, extensions, and the pi-mono monorepo layout. Use when running `pi`, driving it headlessly/programmatically, configuring local models, writing pi extensions/skills, or answering "how does pi do X". Keywords pi, pi.dev, pi coding agent, earendil, pi-mono, llamacpp, models.json, rpc mode, json mode, qwen.

Reference for Claude Code features and best practices: writing CLAUDE.md, skills, hooks, sub-agents, agent teams, headless mode, scheduled tasks, worktrees, batch, frontend verification. Use when authoring/auditing CLAUDE.md, configuring Claude Code, writing skills, or orchestrating agents. Keywords CLAUDE.md, skill, hook, sub-agent, agent team, headless, worktree, /loop.

Design, build, review, or debug article and web demos across the full range from static to live: semantic HTML/CSS layouts, responsive components, design systems, Grid/Flexbox, accessibility AND interactive Canvas 2D, SVG, WebGL/Three.js, physics/mechanical simulations, sliders, scrubbers, requestAnimationFrame loops, React/MDX/Deno/TanStack embeds. Use when building a non-interactive figure or page layout, a live model-driven explainer (Bartosz Ciechanowski style), or anything between. Keywords: demo, graphics, canvas, WebGL, Three.js, animation, simulation, HTML, CSS, layout, Grid, Flexbox, semantic markup, responsive, accessibility, interactive explainer.

Use when explaining, designing, reviewing, or debugging general Temporal concepts from docs.temporal.io, including Temporal Platform architecture, Workflows, Workflow Execution, Activities, Workers, Event History, failure detection, Workflow message passing, Child Workflows, Temporal Service, Namespaces, Task Queues, durable execution, replay, determinism, Signals, Queries, Updates, retries, and timeouts. Fetch current official Temporal docs and inspect embedded page links before giving precise guidance; use SDK-specific skills only when exact language APIs or code semantics are the main task.

Design, build, review, or debug article and web demos across the full range from static to live: semantic HTML/CSS layouts, responsive components, design systems, Grid/Flexbox, accessibility AND interactive Canvas 2D, SVG, WebGL/Three.js, physics/mechanical simulations, sliders, scrubbers, requestAnimationFrame loops, React/MDX/Deno/TanStack embeds. Use when building a non-interactive figure or page layout, a live model-driven explainer (Bartosz Ciechanowski style), or anything between. Keywords: demo, graphics, canvas, WebGL, Three.js, animation, simulation, HTML, CSS, layout, Grid, Flexbox, semantic markup, responsive, accessibility, interactive explainer.

Write, run, debug, package, and make reproducible standalone Python scripts with uv. Use when authoring uv scripts, adding PEP 723 inline metadata, choosing uv run/init/add --script forms, building shebang executables, locking scripts for reproducibility, handling uv cache or CI cache behavior, or picking lightweight script dependencies. Keywords: uv, uv run, uv script, PEP 723, inline metadata, standalone python script, requires-python, uv lock --script, uv add --script, shebang, hook script.

Anton ADR lifecycle — author new architectural decision records, list existing ones by status or affects-category, and mark old decisions superseded. Use when capturing a decision (especially after `cluster-intake-gatekeeper` returns an ADD/DEFER/REJECT verdict), when reviewing prior decisions before changing direction, when checking if a candidate component has been removed before, or when promoting a decision out of memory into a durable record. ADRs live in `context/adrs/` and are immutable — supersession is the only way to change a decision. The ADR index is built by scanning ADR files directly and injected into every Codex session by `.Codex/hooks/inject_adr_index.py`. Keywords — ADR, architecture decision record, decision log, supersede, decision history, why did we, prior decision, recorded decision, MADR, immutable, intake handoff, cluster-intake-gatekeeper handoff, removal graveyard, reverted decision.

Intake gate for adding new system or infrastructure components to Anton. Asks the user to declare intent (concrete need, honest learning, or both), then applies the matching rubric — full production rubric for concrete need, contained-learning rubric for learning intake — and returns add / defer / reject with an ADR-ready summary. Welcomes honest learning intake (anton is partly a learning cluster; "things that don't scale" are okay when declared) but rejects completionism dressed as need. Read-only — never scaffolds manifests, never applies to the cluster. Use when asking "should I add X", "can I run X on the cluster", "is X worth adopting", "I want to try X", "I want to learn X", "evaluate new component", "vet this helm chart", "cluster intake", "new app decision", before scaffolding a new Flux app, or when tempted by a shiny project on HN. Hands passing candidates off to add-flux-app. Keywords — intake, adopt, install, new component, new app, evaluate, should I run, worth it, learning, experiment, try out,

Expose a workload for access. Four paths: envoy-internal (LAN via split-horizon DNS), Tailscale Ingress (internal remote HTTP with browser-trusted TLS), Tailscale Service annotation (raw TCP / non-HTTP), envoy-external + Cloudflare tunnel (genuinely public, requires explicit approval). Handles HTTPRoute authoring, DNSEndpoint for secondary domains, and per-domain cert wiring.

Triage ntfy.sh-routed alerts in Anton — identify which alert fired, why it fired (or why it didn't deliver), and propose a fix. Use when "got an ntfy alert", "alert just fired", "ntfy not delivering", "AlertmanagerClusterFailedToSendAlerts", "AlertmanagerFailedToSendAlerts", "code 40014", "attachments not allowed", "iOS push missing", "test the ntfy receiver", "send a test alert", "what just paged me", "is ntfy working". Combines kube-prometheus-stack Alertmanager API, the self-hosted ntfy server (ADR 0026), and the ntfy CLI for poll/publish probes. Read-only by default; proposes edits the operator applies.

Anton planner skill — author, update, and close multi-session initiatives (migrations, rollouts, long-running refactors) in `context/plans/`. Use when starting a multi-session initiative, tracking next steps on in-flight work, migrating a memory entry to a durable plan, closing a completed initiative, or reviewing what's open. Plans live at `context/plans/NNNN-kebab-slug.md` and are mutable — they capture execution state (what's next, what's blocked, log of decisions made during work) while ADRs capture immutable decisions (why). The active-plan index is built by scanning plan files directly and injected into every Codex session by `.Codex/hooks/inject_plans_index.py`. Keywords — plan, planner, initiative, track work, multi-session, next steps, checklist, migration plan, rollout plan, roadmap, in-flight, blocker, close plan, review-by, exit plan, timebox, memory-to-plan handoff.

Use when Codex needs to inspect or troubleshoot the Anton cluster's self-hosted Temporal deployment with the local Temporal CLI, including checking cluster health, namespaces, workflow visibility, schedules, search attributes, Web UI reachability, or Kubernetes readiness for the `temporal` namespace. Prefer this for Temporal CLI tasks in Anton rather than generic Temporal SDK guidance.

Kubernetes-layer health triage for Anton. Use when Codex checks whether the cluster is OK, Flux is healthy, platform controllers are ready, CNI or DNS is failing, cert-manager or ESO is stuck, gateways are broken, cloudflared is disconnected, or apps are unhealthy.

Remote access reference for Anton. Use when Codex needs to reach or reason about kubectl, flux, talosctl, Tailscale MagicDNS, kubeconfig, talosconfig, off-LAN node access, or failed connectivity to k8s nodes.

Deploy wcygan.net as a pure static site to Cloudflare via Workers Assets. Use when configuring static prerendering, updating vite.config.ts Nitro preset, writing wrangler.jsonc, tuning Cloudflare Workers Builds dashboard settings, switching CI or deploys to Deno, or debugging failed Cloudflare deploys. Keywords cloudflare, wrangler, workers assets, pages, deploy, prerender, static, nitro preset, deno ci, workers builds, wrangler.jsonc, deno.lock, .output/public

Understand and improve site readiness for AI agent discovery — the isitagentready.com checks covering robots.txt, sitemap.xml, Markdown for Agents, MCP server cards, OAuth discovery, Agent Skills, WebMCP, and related emerging standards. Auto-loads when auditing a site against isitagentready.com, deciding which standards apply to a personal blog vs an API-backed service, debugging a missing well-known endpoint, or extending wcygan.net's auto-generation plugin. Keywords isitagentready, agent-ready, agent ready, robots.txt, sitemap.xml, llms.txt, Content-Signal, Markdown for Agents, MCP server card, MCP, OAuth discovery, Agent Skills, WebMCP, Web Bot Auth, well-known, x402, UCP, ACP, site-metadata-plugin, RFC 9309, RFC 8288, RFC 9727, RFC 9728, RFC 8414.

Verify CI and deploy health for wcygan.net — programmatically check that the latest GitHub Actions CI run on main is green, that Cloudflare Workers Builds produced and activated a new deployment, and that the live site passes the regression suite at scripts/verify-prod.sh. Use after pushing to main, when debugging "did my deploy land", when a PR is green but prod looks wrong, when scheduling a release check, or when investigating whether a user-reported issue is a deploy failure or a content bug. Keywords CI status, actions passing, deploy landed, deploy check, cloudflare deploy, workers builds, wrangler deployments, regression test, prod smoke, post-deploy verify, is it broken, release check, github actions, build status, verify prod.

Diagnose and fix GitHub Actions workflow failures, CI/CD pipeline issues, YAML syntax errors, and workflow configuration problems. Use when debugging failed CI runs, fixing workflow errors, or optimizing GitHub Actions. Keywords: github actions, CI, workflow, pipeline, yml, yaml, actions failure, CI failure, workflow error

Scaffold a new MDX blog post for wcygan.net. Creates `src/posts/<slug>.mdx` with correct frontmatter and prepends a matching `<item>` to `public/rss.xml` so the two stay in sync. Use when the user asks to "add a blog post", "new post", "write a post", or similar.

Verify that wcygan.net production is serving the TanStack Start + Vite build (not a cached SvelteKit deploy or stale worker). Use when debugging "did my deploy land", comparing local vs prod output, confirming a rollback worked, investigating stale content after a push, or diagnosing hydration/routing failures that could be caused by mixed versions. Keywords verify, fingerprint, deploy check, prod, production, tanstack vs sveltekit, __TSR, cache, stale, assets, hydration, rollback verification, which version

Design system and UI/UX conventions for wcygan.net — a TanStack Start + Tailwind v3 personal blog with a minimal, readable, editorial aesthetic. Auto-loads when working in /Users/wcygan/Development/wcygan.net, editing src/styles/app.css, src/routes/__root.tsx, or any route/component/MDX post on that site. Use when styling pages, adding routes, writing MDX posts, touching the header/nav/footer chrome, or changing colors, type, spacing, or layout. Keywords: wcygan.net, wcygan blog, personal site design, minimal blog design, system fonts, primary blue, link blue, cornflower blue, bio-highlight banner, post-list, blog-post, 800px container, MDX frontmatter, Mermaid diagram, Shiki, Tailwind v3, TanStack Start blog, __root chrome, design.md, app.css

Project stack guide for wcygan.net covering Deno runtime, TanStack Start framework, MDX markdown rendering, and Mermaid diagram components. Auto-loads when working on routes, components, blog posts, MDX config, Vite config, mermaid diagrams, or build/deploy. Keywords: deno, tanstack, tanstack start, tanstack router, mdx, markdown, mermaid, diagram, vite, nitro, blog post, route, shiki

Fan out a 12-persona review panel against will_cygan_resume.typ in parallel, with work-experience/*.md supplied as the source-of-truth evidence base, and produce a consolidated dashboard with consensus verdict, dissent highlights, and deduplicated top issues. Use when the user wants a comprehensive multi-angle resume review, asks for a "panel review", "full review", "multi-angle review", or wants every reviewer to weigh in at once. Covers ATS parsing, recruiter triage, hiring manager, technical screener, bar raiser, staff IC peer, leveling committee, skip-level exec, domain SME, typography, future-self skeptic, and career coach perspectives.

Structured 3-round debate about a single resume bullet. Runs hiring-manager, bar-raiser, and future-self-skeptic through opening critique, rebuttal, and rewrite rounds. Produces either a consensus rewrite or an articulated disagreement for the user to adjudicate. Accepts either the bullet text or a line number as $ARGUMENTS. Use when the user is uncertain about a specific bullet, asks "is this bullet good?", wants multiple perspectives on one line, or says a reviewer flagged a specific bullet.

Generate a ranked interview prep sheet from the resume. Three interviewer personas (hiring-manager, bar-raiser, domain-sme-systems) each generate 5 questions they would ask based on specific resume bullets. The questions are then deduplicated and ranked by difficulty tier (softball, standard, hard, trap). Includes a coverage audit flagging bullets no persona asked about. Optionally accepts a job description as $ARGUMENTS to target the questions for a specific role. Use when the user has an interview coming up, asks for interview prep, or wants to rehearse what they'd be asked about their resume.

Optimize engineering resumes using proven STAR/XYZ methodologies, ATS best practices, and hiring manager insights. Use when reviewing resumes, improving bullet points, tailoring to job descriptions, or enhancing professional presentation. Keywords: resume, CV, bullet points, STAR, XYZ, ATS, job description, optimize, tailor, action verbs, quantify, achievements

Run a section-focused resume review panel. Routes to a curated subset of the 12 reviewer sub-agents based on the section argument (work-experience, skills, projects, formatting, narrative, or header). Faster and more targeted than a full panel — uses 3-6 relevant personas instead of all 12. Use when the user wants to review only a specific part of their resume, has just edited one section, or asks about one area like "the skills section" or "the formatting" or "my bullets".

ATS text-extraction regression gate for the resume repo. Runs the compiled PDF through three open-source parsers (pdftotext, pdftotext -layout, Apache Tika) that real ATSs actually use, asserts clean extraction across nine structural checks, and surfaces disagreement between parsers as diagnostic signal. Use when the user asks about ATS parseability, text extraction, PDF parsing reliability, whether a resume variant breaks for ATSs, how the extraction gate works, running or debugging extraction-check, comparing extractor outputs, adding assertions, writing new broken fixtures, interpreting cross-extractor disagreement, or the text-extraction hypothesis at .claude/context/text-extraction-hypothesis.md. Keywords ATS, applicant tracking system, text extraction, PDF parsing, pdftotext, poppler, tika, Apache Tika, reading order, mojibake, cross-extractor, extraction check, regression gate, resume parser, parseability, soft hyphen, url dedup.

Vendor @preview Typst packages into template/ or bootstrap new in-house Typst templates from scratch. Use when the user asks to vendor, fork, copy, pin, localize, or maintain a Typst template; when a @preview package breaks on a newer typst version; when creating a new template.typ from scratch; or says things like "vendor modern-cv", "fork this typst package", "bring this template in-repo", "write our own resume template". Keywords typst, vendor, fork, @preview, template, modern-cv, fontawesome, linguify, typst.toml, lib.typ, MIT attribution, version drift.

Before/after resume review. Compares working-tree will_cygan_resume.typ to a previous version (defaults to git HEAD, accepts any git ref or file path as $ARGUMENTS) and runs a 5-persona panel on both versions. Reports which issues the user's edits resolved, which remain, and which are new. Use when the user has just made edits and wants to know whether they improved the resume or introduced new problems, or when comparing two versions of the resume.

Expert understanding of raft-a-tui's dual-thread architecture (Raft Ready loop + ratatui TUI). Refactor and optimize integration between consensus layer and terminal UI. Use when working with TUI rendering, event loops, Ready loop patterns, channel communication, or performance optimization. Keywords: TUI, ratatui, event loop, Ready loop, refactor, integration, channels, state updates, performance, real-time, rendering

Build terminal user interfaces using ratatui - widgets, layouts, event handling, rendering, and application patterns. Use when creating TUIs, implementing terminal UIs, building interactive CLI apps, or working with ratatui/crossterm. Keywords: ratatui, TUI, terminal UI, crossterm, widgets, layout, event handling, immediate mode rendering

Expert guidance for implementing Raft consensus algorithm using raft-rs (TiKV). Covers RawNode integration, Ready loop patterns, storage/network layers, debugging elections and log replication, and production deployment. Use when implementing Raft, debugging consensus issues, or designing distributed systems. Keywords: raft, consensus, distributed systems, raft-rs, TiKV, RawNode, Ready loop, election, log replication, quorum, leader election, state machine

Create new Rust crates in this Cargo workspace following established patterns for dependency management, testing, documentation, and project structure. Use when adding new workspace members, creating modules, or structuring code. Keywords: cargo, crate, workspace, module, new crate, add crate, dependency, Rust project

Analyze and reduce cognitive load in code by identifying complex conditionals, deep nesting, shallow modules, excessive abstractions, and architectural over-engineering. Use when reviewing code quality, refactoring for maintainability, or simplifying systems. Keywords: cognitive load, complexity, refactor, simplify, maintainability, readability, mental overhead, abstraction, architecture review

Understand the toy relational database architecture, crate organization, data flow, and implementation patterns. Use when exploring the database codebase, understanding query execution, debugging storage/indexing, or planning new features. Keywords: sql-database, database, crate, storage, executor, planner, parser, catalog, WAL, buffer pool, index, architecture