Jeremylongworth Source@AgentSkills

Review environment configuration for apps, services, CI, deployment, and runtime settings. Use when Codex is asked to inspect env vars, config files, secrets, feature flags, environment parity, config drift, or deployment-specific settings.

Design reusable prompt regression tests for agent skills and workflows. Use when Codex is asked to prevent a skill from regressing, create prompt test cases, cover known failures, or define repeatable evaluation prompts.

Design, audit, and improve sales pipeline stages, qualification, CRM hygiene, forecasting, deal reviews, pipeline coverage, stage exit criteria, sales metrics, opportunity management, and revenue operating cadence. Use when Codex is asked about pipeline health, sales process, forecast accuracy, deal stages, CRM fields, sales dashboards, or funnel bottlenecks.

Prepare review-only synthetic character bibles for AI-assisted creator personas, mascots, recurring fictional hosts, virtual influencers, or synthetic spokescharacters. Use when Codex is asked to define a synthetic character's identity, voice, visuals, backstory, boundaries, disclosure rules, likeness and voice risks, provenance notes, prohibited claims, continuity, or production guardrails.

Turn skill, bundle, feature, or workflow goals into testable acceptance criteria. Use when Codex is asked to define pass/fail checks, map requirements to verification steps, or make a skill evaluation measurable.

Prepare review-only moderation and escalation briefs for creator communities. Use when Codex is asked to review comments, DMs, member posts, Discord or community issues, potential rule violations, harassment, spam, inflammatory threads, removal questions, response options, reviewer notes, or moderation boundary decisions for a creator, influencer, manager, or community team.

Plan, run, analyze, and review game playtests, usability sessions, player feedback, telemetry, onboarding tests, difficulty tests, controls tests, fun/enjoyability research, prototype validation, and playtest reports. Use when Codex is asked to test whether a game is understandable, fun, balanced, usable, accessible, or ready for release.

Package, test, publish, price, and update browser-playable HTML5 games on itch.io. Use when Codex is asked about itch.io HTML5 uploads, ZIP requirements, index.html, iframe embeds, fullscreen mode, mobile-friendly settings, case-sensitive paths, HTTPS assets, compression, game page setup, donations, pay-what-you-want pricing, creator payments, bonus files, or itch.io launch readiness.

Audit and improve owned social profiles such as X, Facebook, LinkedIn, YouTube, Instagram, TikTok, and company pages. Use when Codex is asked to review profile positioning, bios, links, pinned or featured posts, profile images, banners, public visibility, verification state, cross-platform consistency, or profile conversion paths.

Design, audit, and test pricing and packaging strategy. Use when Codex is asked about SaaS pricing, product tiers, packaging, value metrics, subscriptions, usage-based billing, freemium, trials, discounts, pay-what-you-want, donations, monetization, willingness to pay, pricing pages, billing implications, or pricing experiments.

Plan, write, and review ethical B2B/B2C prospecting and outbound outreach. Use when Codex is asked about prospect lists, ICP targeting, cold email, cold calls, LinkedIn/social selling, outreach sequences, personalization, lead qualification, reply handling, objection replies, deliverability, consent, or compliant outbound workflows.

Evaluate, forward-test, and improve Codex skills. Use when Codex is asked to test a skill, compare performance with and without a skill, improve a skillset, make skills stronger, validate skill usefulness, create acceptance tests for skills, or iterate skill instructions based on realistic scenarios.

Research, compare, and synthesize markets, categories, competitors, alternatives, pricing, feature sets, audience segments, positioning, review themes, and strategic opportunities. Use when Codex is asked for competitive analysis, market maps, category research, competitor messaging, pricing comparison, SWOT, positioning gaps, or market-entry research.

Review whether an app, service, feature, or workflow is ready for production. Use when Codex is asked for launch readiness, operational readiness, release gates, observability, support, security, rollback, or go/no-go assessment.

Improve sales discovery, demos, proposals, business cases, objection handling, negotiation, mutual action plans, procurement, handoffs, and sales enablement assets. Use when Codex is asked to help convert opportunities, write sales collateral, create demo flows, prepare call plans, handle objections, or improve close rates.

Review the context, token, tool, and process overhead added by agent skills. Use when Codex is asked whether a skill is too verbose, too broad, costly to invoke, duplicative, or worth its context/tool overhead.

Report creator affiliate performance from supplied tracking data. Use when Codex is asked to analyze affiliate links, codes, clicks, conversions, revenue, commissions, refunds, EPC, conversion rate, attribution caveats, cookie windows, sponsor reporting, owned-product reporting, or creator commerce performance.

Create creator production operations briefs for budgets, assets, assistant or contractor handoffs, capacity, dependencies, and weekly production flow. Use when Codex is asked to organize influencer production work, creator team tasks, editor handoffs, asset checklists, production budgets, publishing dependencies, contractor briefs, assistant instructions, or production bottlenecks.

Plan, conduct, synthesize, and review customer research and demand validation. Use when Codex is asked about customer interviews, surveys, usability studies, JTBD, buyer discovery, problem validation, demand validation, research scripts, research synthesis, voice-of-customer analysis, customer evidence, or research-backed product/marketing decisions.

Design, evaluate, and document experiments, A/B tests, product tests, growth tests, validation tests, hypotheses, primary metrics, guardrails, sample considerations, decision rules, and learning plans. Use when Codex is asked to test an idea, validate a change, run an experiment, compare variants, or create an experimentation plan.

Compose, generate, direct, implement, and review game music and soundtracks. Use when Codex is asked about themes, motifs, adaptive music, interactive scores, soundtrack generation, stems, loops, transitions, vertical layering, horizontal resequencing, combat music, exploration music, boss themes, ambience beds, licensed music, music prompts, FMOD, Wwise, Unreal MetaSounds, or Unity music systems.

Coordinate sales and marketing strategy across positioning, ICP, funnel, pipeline, channels, lifecycle, analytics, experiments, and revenue operations. Use when Codex is asked to create a GTM plan, growth strategy, launch plan, revenue plan, marketing/sales operating system, funnel audit, or cross-functional sales-and-marketing program.

Facilitate "grill me" style guided discovery interviews that ask the user focused questions, progressively elicit missing requirements, and build an aligned draft artifact such as a BRD, PRD, feature spec, implementation brief, decision memo, or project plan. Use when Codex should interview the user, clarify ambiguous goals, gather stakeholder context, fill requirement gaps, keep the user aligned during planning, or turn partial ideas into structured requirements through an interactive question flow.

Design, audit, and improve accessibility and inclusive user experiences. Use when Codex is asked about WCAG, keyboard navigation, focus states, screen readers, ARIA, contrast, captions, reduced motion, forms, semantic HTML, accessible UI, game accessibility, inclusive design, accessibility QA, or compliance-sensitive accessibility work.

Set up AgentSkills in a new or existing repository in a focused way. Use when Codex is asked to onboard AgentSkills, choose the right skillset or skills for a repo, create or update project agent instructions, decide project vs user scope, configure focused routing, recommend MCP presets, or verify that AgentSkills works in a repository without installing unnecessary broad bundles.

Plan analytics instrumentation for product, web, backend, funnel, and business events. Use when Codex is asked to define events, properties, user/account identity, tracking specs, validation, QA, or data quality for analytics.

Create, review, and refine Business Requirements Documents (BRDs), business requirements, stakeholder requirements, scope statements, current/future-state workflows, business rules, assumptions, constraints, success criteria, traceability, and approval-ready requirement packages. Use when Codex is asked to write a BRD, gather business requirements, convert discovery notes into business requirements, review requirement completeness, align stakeholders, or prepare business-analysis documentation before solution design.

Prepare review-only reputation risk memos for creator content, campaigns, public replies, sponsorships, AI-assisted content, rights questions, disclosure checks, brand-safety concerns, platform-policy uncertainty, and controversy response planning. Use when Codex is asked to assess influencer reputation risk, sponsored post risk, AI labeling, copyright or usage-risk flags, public backlash, apology or response options, or hold/revise/go recommendations for a creator, manager, or creator team.

Design, storyboard, implement, render, and review game cutscenes, intro videos, cinematics, trailers, and in-engine camera sequences. Use when Codex is asked about storyboards, animatics, shot lists, blocking, camera language, Sequencer, Unity Timeline, Cinemachine, Unreal cinematics, Blender storyboarding, intro movies, prerendered videos, gameplay-to-cutscene transitions, cinematic trailers, subtitles, timing, editing, or cinematic production notes.

Define ICP, buyer personas, positioning, category strategy, messaging, value propositions, offers, campaign strategy, brand narrative, launch strategy, and go-to-market messaging. Use when Codex is asked to clarify who to sell to, why they buy, how to differentiate, or how to package a campaign or offer.

Explore intent and shape ideas before implementation planning. Use when Codex is asked to brainstorm, ideate, plan a feature/product/workflow, clarify a vague idea, compare approaches, discover requirements, create a PRD, define MVP scope, or turn rough goals into an implementation-ready brief.

Design, implement, debug, and review player social interaction systems for games. Use when Codex is asked about chat, voice, friends, parties, lobbies, guilds, clans, presence, invites, matchmaking social flow, emotes, pings, reputation, blocking, muting, reporting, moderation, privacy, parental controls, user-generated content, community safety, trust and safety, or social UX.

Define, audit, and improve marketing analytics, GA4 events, conversion goals, attribution, dashboards, funnel metrics, CAC, LTV, campaign measurement, experiment design, UTM governance, CRM-source reporting, and data quality checks. Use when Codex is asked to measure sales/marketing performance or diagnose reporting.

Plan creator-owned product launches. Use when Codex is asked to launch a creator digital product, course, merch drop, paid community, membership, bundle, preorder, waitlist, offer page, content sequence, email sequence, fulfillment plan, support plan, or product launch calendar.

Design, audit, and improve product analytics instrumentation, event taxonomy, telemetry schemas, funnels, activation metrics, retention metrics, tracking plans, observability events, QA, and data quality. Use when Codex is asked about product analytics events, app/game telemetry, funnel instrumentation, analytics implementation, event naming, properties, user identity, or measurement-ready product design.

Review creator content performance from supplied analytics. Use when Codex is asked to analyze influencer posts, videos, Reels, Shorts, TikToks, newsletters, podcasts, watch time, retention curves, hooks, titles, thumbnails, audience response, top content, weak content, or next creator content actions.

Audit, diagnose, and optimize React and Next.js application performance. Use when Codex is asked about Core Web Vitals, slow interactions, bundle size, caching, Server Components, Client Component boundaries, re-renders, memoization, image/font/script optimization, lazy loading, route performance, production readiness, or React/Next.js performance regressions.