robertguss

Refresh a repo's documentation for a live production system — audit stale docs, rebuild a focused doc set with diagrams, update AI-agent knowledge bases, and improve code-level docstrings

Full-pipeline iOS App Store opportunity research. Discovers underserved niches, analyzes competitor gaps, estimates revenue, produces scored top-3 opportunity reports, and writes MVP PRDs — all through browser and web research. Use when the user wants to find…

Generate platform-specific, actionable growth playbooks for mobile apps. Use when planning a Product Hunt launch, creating TikTok/Reels content strategies, setting up Apple Search Ads campaigns, preparing App Store featuring submissions, building referral…

Systematically analyze app reviews for competitor research or own-app reputation management. Categorize reviews by sentiment and themes (bugs, missing features, UX complaints, pricing objections, praise, support issues), identify actionable patterns, produce…

Optimize iOS App Store and Google Play Store listings for maximum discoverability and conversion. Perform competitive keyword research, craft keyword-optimized titles/subtitles/descriptions, design screenshot sequences, and generate A/B test variants. Use…

Orchestrate iOS/macOS app scaffolding and optional skill adoption for existing projects. Use when users want a guided wizard that can scaffold with XcodeGen and optionally install xcode-makefiles and simple-tasks.

Install a fast local task workflow for single-project planning with `scripts/task.sh` (claim, done, status, reporting) backed by `tasks/TASKS.md` and optional `tasks/details/` notes. Use for lightweight in-progress task coordination, not full team issue…

Install strict Xcode Makefile tooling for iOS/macOS projects, including build/run/test scripts with AGENT_NAME-based per-agent isolation under build/. Use when a project needs reproducible local CLI builds without full app scaffolding.

Opt-in council rung for a Mycelium idea: commission identical prompts across N models, collect per-model reports, and reconcile with retained dissent. Suggest only on v1 replication triggers. Use when program/packs/council/ is present. Does not invent…

Cheap second-opinion rung for a Mycelium idea: one different model, identical commissioning prompt, exactly one model-report, no reconciliation. Use when program/packs/council/ is present. Does not invent council/ladder/replicate verbs. Does not commit unless…

Mark a Mycelium idea clarified: destination reached, handoff packet buildable. Use when the human says the idea is decided enough to hand off, or asks to clarify. Does not write the packet.

Write the Mycelium implementation packet and set state to handed-off. Use when the idea is clarified and the human is ready to implement. Legal only from clarified. Does not reopen locked decisions.

Operate a Mycelium idea instance with the mycelium CLI: scaffold, generate registered artifacts, check conformance, change tier, lifecycle transitions, wake, handoff, index, status, publish, and supersede. Use when working inside an idea repo that has…

Cross-idea Mycelium portfolio view: run status --all, interpret local-only partials honestly, and do not mutate. Use when surveying many ideas or the human asks for portfolio status. Pass --offline when hermetic or gh is absent.

Choose the Mycelium ritual from the human's goal: new spark, work a live idea, park, wake, research program, second opinion, council, clarify, or handoff. Use at the start of a sitting in a mycelium.toml repo, or when the human states a goal but not a command.

Park a Mycelium idea on purpose: exploring → simmering with a required revisit date or event. Use when the human wants to stop, come back later, or says park/simmer. Not for blocked work and not for unstateable fog.

Stewards Conveyor repo documentation so roadmap, handoff, changelog, ADRs, backlog docs, and domain language stay consistent without doc churn. Use when work changes Conveyor product direction, artifact boundaries, roadmap phase, session handoff state,…

Agent eXperience Interface (AXI) — ergonomic standards for building CLI tools that agents use via shell execution. Use when building, modifying, or reviewing any agent-facing CLI.

Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes…

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or…

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and…

Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase…

Sets up an `## Agent skills` block in AGENTS.md/CLAUDE.md and `docs/agents/` so the engineering skills know this repo's backlog backend (GitHub or local markdown), triage label vocabulary, and domain doc layout. Run before first use of `to-issues`, `to-prd`,…

Set up a fresh clone of the Next.js + Convex + Clerk Web App Starter Kit through a running, verified development app. Use for first-time setup, missing CLERK_JWT_ISSUER_DOMAIN errors, or broken Clerk-to-Convex authentication.

Set up or extend Clerk authentication and owner authorization in this Next.js and Convex starter.

Operate the Clerk CLI (`clerk` binary) for authentication, user/org/session management, impersonation, local webhook testing, deploy verification, instance config, env keys, feature toggles, and any Clerk Backend, Platform, or Frontend API call. Use when the…

Add Clerk authentication to any project by following the official quickstart guides.

Manages shadcn components and projects — adding, searching, fixing, debugging, styling, and composing UI, including chat interfaces. Provides project context, component docs, and usage examples. Applies when working with shadcn/ui, component registries,…

Set up Convex authentication with proper user management, identity mapping, and access control patterns. Use when implementing auth flows.

Guide to using Convex components for feature encapsulation. Learn about sibling components, creating your own, and when to use components vs monolithic code.

Discover and use convex-helpers utilities for relationships, filtering, sessions, custom functions, and more. Use when you need pre-built Convex patterns.

Set up a fresh clone of the TanStack Start + Convex + Clerk starter kit from dependencies to a running dev server. Handles Convex dev, Clerk app/JWT setup, .env.local, and the CLERK_JWT_ISSUER_DOMAIN Convex env var. Use when the user says "set up this starter…

Set up or extend authentication patterns in this Convex + Clerk project

Create Convex functions (queries, mutations, actions) with proper validation, auth, and patterns for this project

Quick reference for convex-helpers utilities available in this project - custom functions, relationships, validators, and more

Guide for safely migrating Convex schema changes - adding fields, changing types, renaming, and batch processing

Scaffold a new feature module in this Convex project - creates schema, functions, and frontend integration

Design Convex database schemas with proper validation, indexes, and relationship patterns

Run uv/ruff/ty/pytest quality gates for python-foundry-template

Add a CLI command to python-foundry-template

Secrets via fnox + age for python-foundry-template (no dotenv secret storage)

Orchestrate an artifact-driven research program: discovery interview (one question at a time with recommendations), resume from Git artifacts, rigor tier and research-graph design, Blueprint/Charter guidance, and next legal stage selection. Use when the user…

Produce the just-in-time five-item stage package for an artifact-driven research program: canonical stage prompt, repository installation task, attachment manifest, fresh-session launch message, and post-stage validation task with recommended commit message.…

Run the independent validation gate for artifact-driven research program outputs: required sections, metadata, identifiers, citations, evidence ledgers, checklists, scope, authority, placeholders, and manifest consistency. Use after a stage produces an…

Add a new kit tool end-to-end (core → CLI → API → C16 tests → just check). Use for any new tool id.

Secrets, age keys, OAuth/break-glass, and student-PII hard boundaries for the kit.

Change CLI flags, API routes, or SPA surfaces without moving pipeline logic out of kit-core.

Drive Elara's Mix CLI, persistent chat, and Rust TUI the way a user does. Use when proving ask/chat/TUI/session/tool behavior, after changing those surfaces, or before claiming a user-visible fix works.

Builds, extends, and verifies Surface intent-driven interfaces while preserving semantic parity across React, HTML, JSON, Markdown, and actions. Use when authoring a Surface brief or spec, selecting semantic components, changing the catalog or compiler,…

Personal financial advisor that runs the Financial Order of Operations on a local ledger. Use whenever the user wants money advice, a snapshot, a plan, a budget check, debt payoff order, or help with accounts.yaml and transaction CSVs. Load this skill before…

Use the probe CLI to spike HTTP APIs non-interactively, save redacted exchanges, and promote them into a personal catalog. Prefer probe over throwaway Python/curl scripts when exploring authenticated APIs.