relay

Drive the permanent fresh-install staging environment — a customer-identical `npx orionfold-relay` first-run against an isolated data dir. Use when the user mentions staging, fresh-install testing, customer-experience simulation, "spin up staging", "launch the staging instance", tear down staging, isolated data-dir launch, reproducing a customer report, or seeding/clearing staging scenarios. Also triggers on "relay-staging", "staging harness", "customer first-run", "PLG-S", "simulate a customer install", or any request to stand up a clean isolated Relay instance the operator can watch. This is the SUBSTRATE the staging-cli-run / staging-browser-smoke / staging-evaluate skills ride on; invoke it first to prepare a running instance. Do NOT use for dogfooding from a git worktree (use worktree-production) or for the ephemeral CI smoke.

Technical architecture advisor that maintains Technical Decision Records (TDRs) and advises on system design, integration patterns, architectural changes, and drift detection. Use this skill when the user asks about architecture review, system design, blast radius analysis, integration design, technical decisions, pattern consistency, coupling analysis, technical debt assessment, TDR management, architecture drift, or pattern drift. Also triggers on "architecture review", "blast radius", "impact analysis", "integration design", "TDR", "technical decision", "pattern audit", "coupling", "tech debt", "architecture health", "how should this integrate", "what's the architectural impact", "what breaks if I change", "document this decision", or any question about system-level design, cross-cutting concerns, architectural trade-offs, or whether the codebase is drifting from its own conventions. Do NOT use for building components, UX review, writing feature specs (use product-manager), running tests (use quality-manag

Scaffold an ainative-native app by composing existing primitives (agent profiles, workflow blueprints, user tables, schedules, document routing) from a YAML manifest instead of writing TypeScript. Use when the user asks to "create an ainative app", "scaffold an app", "build an app manifest", "design an app in ainative", "wealth-manager-style app", "app from config", "compose an ainative app", or when they want to package a workflow + profile + schedule + table combination as a reusable bundle. Also triggers when the user wants to turn an ad-hoc workflow into a shippable app or when a proposed feature can be built entirely by composition of existing primitives. Do NOT use for writing new TypeScript components, new DB migrations, new chat tools, or net-new primitive kinds — those are outside the config-over-code contract; route to `product-manager` or `architect` instead.

Transform raw product ideas into structured, implementable feature specs. Use this skill whenever the user mentions grooming features, creating a roadmap, extracting features from ideas, updating feature status, checking what to build next, prioritizing work, maintaining a changelog, or managing product backlog. Also triggers on "what should we build", "groom", "feature spec", "roadmap update", "mark feature as", "what's the status", "prioritize", or any request to turn ideas into actionable development tasks. If the user has an ideas/ directory and wants to organize it into buildable units, use this skill.

Meta-orchestrator that reads project state holistically to recommend what to work on next. Use this skill when the user asks "what should I work on next", "project health", "status overview", "plan a sprint", "are we on track", "vision check", "retro", "lessons learned", "what's the priority", or any question about overall project direction, velocity, or process improvement. Do NOT use for writing feature specs (use product-manager), building components directly, running tests (use quality-manager), or UX review.

Sensitive-material analyst intended only for an operator-verified, locally hosted Ollama endpoint with network tools disabled

Quality assurance orchestrator for testing strategy, test creation, coverage tracking, code review, regression guarding, and browser-based feature evaluation. Use this skill when the user mentions setting up testing, writing tests, checking coverage, code review, quality audit, regression check, "safe to ship", evaluating features in browser, verifying acceptance criteria, or testing strategy. Also triggers on "set up vitest", "write tests for", "what's our coverage", "quality check", "run tests", "evaluate features", "verify in browser", "test the UI", or any request involving test infrastructure, test creation, or quality assurance workflows. Do NOT use for building components, creating feature specs (use product-manager), or UX review.

Balanced general-purpose agent for a wide range of tasks

Project-level overrides for the superpowers writing-plans skill. Adds scope challenge step and required plan sections to every implementation plan.

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

Curates web-design assets into a clean gallery inventory

Drafts structured website section rows for the static-site publisher

Project-level overrides for the superpowers brainstorming skill. Adds scope modes (EXPAND/HOLD/REDUCE) and required design artifacts to every brainstorming session.

Capture web content and save as LLM-friendly markdown. Handles documentation sites (multi-page with search index), articles, blog posts, research papers, and PDFs. Supports custom target folders like ideas/. Triggers on "capture", "scrape", "save this article", "download docs", "grab this post".

Security-focused code review with 2-pass model. Pass 1 blocks merge (SQL safety, race conditions, LLM trust boundary, security). Pass 2 is informational (dead code, magic numbers, test gaps). Use when reviewing code changes, PRs, or before shipping features. Triggers on 'review code', 'code review', 'check this code', 'is this safe to merge'.

Project-level overrides for the commit-commands commit-push-pr skill. Adds auto version-bump heuristic and bisectable commit guidance.

Track and audit deliberate shortcuts left in the code as `DEFER:` markers, and harvest them into a debt ledger that flags any deferral without an upgrade trigger as "later means never" rot. Use when the user asks to review tech debt, audit shortcuts/deferrals, run a defer sweep, check the debt ledger, find untriggered deferrals, or after a deliberately-minimal implementation to record what was left out. Triggers on "defer ledger", "debt ledger", "audit deferrals", "what did we punt", "tech-debt sweep", "DEFER markers", "untriggered shortcuts", "later means never".

Look up documentation from captured reference libraries. Use when the user wants to check docs, look up an API, find how something works, or reference captured documentation. Triggers on phrases like "refer to docs", "look up", "check the docs", "what does the SDK say about", "how do I", "find in docs", "reference docs".

Walk a running Relay staging instance through the ICP journeys (J0–J7 agency-owner + JS1–JS6 solo-founder + JB1–JB3 app-builder) in a live browser the operator watches, capturing per-screen PNGs + console + network logs + a house-format README into a dated output/staging/<date>/ bundle (Mode B). Journeys partition into six session-sized runs (R1–R6) — default to ONE run per invocation. Use when the user mentions the browser walkthrough, the ICP journeys, a Mode B capture, walking the app as an agency owner / solo founder / app builder, screenshotting the first-use flow, capturing the web license-activation ceremony, or producing a reviewable UI bundle. Also triggers on "staging-browser-smoke", "browser smoke", "walk the journeys", "J0-J7", "run R1"–"run R6", "Mode B", or any request to drive the running staging instance through the app and capture what a customer sees. This RIDES on the `relay-staging` substrate — invoke that skill's `setup` + `launch` first (the instance must be up on :3199). Do NOT use for

Record the Relay CLI first-run + free→paid fulfilment journey (Mode A + C) as a GIF and grep-able logs, against the isolated staging install. Use when the user mentions recording the CLI first-run, capturing the fulfilment/activation ceremony, a "cli-first-run GIF", Mode A or Mode C capture, the D4 perpetual- fallback proof, minting a staging license, or producing a terminal recording of `relay license add` → `relay pack add`. Also triggers on "staging-cli-run", "record the CLI", "fulfilment GIF", "D4 proof", or any request to capture the license-activation + pack-install ceremony as a reviewable artifact. This RIDES on the `relay-staging` substrate — invoke that skill's `setup` first (the install must exist). Do NOT use for the browser walkthrough (use staging-browser-smoke) or for evaluating a captured bundle (use staging-evaluate).

Turn a captured staging bundle (output/staging/<date>/) into a verified, ranked dogfooding backlog: features/fix-*.md specs + _IDEAS/backlog.md entries + FILED GitHub issues. Use when the user mentions evaluating a staging bundle, grooming staging findings, Mode D, verify-before-groom, turning a walkthrough/CLI-run capture into fix specs, replicating a customer report, or filing issues from a captured run. Also triggers on "staging-evaluate", "evaluate the bundle", "groom the findings", "Mode D", or any request to convert a captured staging run into reviewable work. This CONSUMES bundles produced by staging-cli-run (Mode A+C) and staging-browser-smoke (Mode B) — run one of those (and the relay-staging substrate) first. Do NOT use to capture a run (use staging-cli-run / staging-browser-smoke) or to stand up the instance (use relay-staging).

Operator drives a real customer first-run (npx orionfold-relay → license → pack → run a workflow) in their own terminal + browser while Codex WATCHES and LOGS proactively — guiding the operator on the exact commands (npx, license activation, pack install) and capturing every bug/UX finding to a session findings file as they surface. Use when the operator says they want to "self-run the first user experience", "walk through as a customer", "I'll drive, you watch", "help me activate the license", "guide me through npx", "operator runs and provides feedback", "tear down the staging folder/server", or any session where the HUMAN interacts with a live instance and CC's job is to narrate + log (and drive the cleanup at wrap), not to drive the clicks. This is the operator-driven complement to the CC-driven staging skills. Opens with a MODE CHOICE (Step 0.5, multi-select): Mode A = local staging mirror on :3199 (your working tree, frozen build), Mode B = the literal npx published package on :3200 (the recommended def

Manage a production-like ainative instance running in a git worktree for dogfooding. Use when the user mentions worktree setup, syncing worktrees with main, running ainative in production mode, dogfooding, isolated DB for worktrees, worktree environment setup, migration sequencing from feature branches, or seeding a worktree instance. Also triggers on "worktree production", "dogfood", "sync worktree", "worktree setup", "production instance", or any request to run the app from a worktree with its own database.

The CRM owner-operator — captures leads, reconciles duplicates, and moves each lead through its two-axis lifecycle from first touch to champion

The intake quality gate — screens a raw lead against six gates before it enters the book, and audits the existing book for hygiene and guardrail violations

The pre-send compliance gate — returns go/fix/no-go on a single 1:1 outreach against channel routing, CAN-SPAM, GDPR, and the cadence cap

The paid-side strategist — recommends where the next ad dollar goes against CAC and ROAS, with explicit stop-loss and scale-up thresholds; advisory only

The organic-first campaign engine — composes, schedules, and measures a campaign from plan to published to results, setting the UTM that leads attribute back to

The content inventory and repurpose planner — tracks source assets, decides what to repurpose next, and writes the channel brief for each creative

Test-only CRM rep for the bundle smoke fixture.

Test-only social writer for the bundle smoke fixture.

Agency account manager who owns client relationships, tracks client health, and writes the per-client status update

Agency bookkeeper for receipt and invoice capture, expense coding, per-client allocation, and budget reconciliation

Agency intake coordinator who reads a new request, classifies the service it needs, and routes it to the right work

Agency new-business researcher who researches prospects and qualifies them before the agency invests in a proposal

Agency-operations runner that onboards new clients — selecting profiles, configuring guardrails, and scheduling deliverables

Agency proposal writer who turns a qualified prospect brief into a scope of work, a proposal, and a ready-to-sign engagement letter

Commercial-real-estate analyst for lease abstraction, rent-roll construction, and comparable research with confidence scoring

Writes commercial-real-estate listing copy and offering one-pagers grounded in submarket comps and defensible claims

Deep CRE renewal methodology — critical-date management, escalation and CAM economics, renewal-option analysis, comp normalization, LOI drafting, and portfolio rent-roll rollup