product-skills

Audit and improve traditional search ranking: on-page SEO, schema markup, entity recognition, and content structure. Use for "audit my SEO", "improve search rankings", "implement schema markup", or "fix my site structure". For AI-visibility auditing and agentfacts, use geo-optimizer.

Audit and improve how a site is represented to AI search and agents. Use for "audit for AI visibility", "optimize for ChatGPT", "check GEO readiness", "analyze hedge density", "generate agentfacts", or "test how LLMs describe my site". Ships a runnable auditor and an agentfacts schema. For traditional search ranking and schema markup, use ai-seo-optimization.

This skill should be used when the user asks to "audit my SaaS", "check if I'm ready to launch", "review my launch checklist", "verify my pricing", "audit my payment setup", "check my AI visibility", "prepare for Product Hunt", "validate my SaaS for launch", or mentions launching a SaaS product. Provides a comprehensive, repeatable checklist with PASS/FAIL verification and actionable next steps.

This skill should be used when the user asks to close the loop on a campaign, define a readback metric, set up analytics readback, decide whether a marketing change actually worked, or promote-or-reject a marketing experiment. Trigger phrases include "close the loop on this campaign," "define the readback metric," "did this marketing actually work," "measure this before shipping," "set up analytics readback," and "promote-or-reject this experiment." Enforces a mandatory discipline across all marketing-skills output: every deliverable must declare, before it ships, the metric it will be judged on, the analytics source that supplies it, the decision rule that separates promote from reject, and the rollback rule that governs when to revert — then route measurement to experiment-stats (A/B tests) or content-scorer (content quality) and record the verdict. Use this whenever a marketing skill's output would otherwise be marked "done" without evidence it moved a real number.

This skill should be used when the user asks to "score this marketing copy", "is this content good enough to ship", "recursively improve this until it scores 90+", "check against rejected patterns", "run a content quality gate", or wants deterministic, repeatable scoring of marketing or product copy against a rubric that remembers what got rejected before. Use it any time copy needs a PASS/REVISE verdict before shipping.

This skill should be used when the user asks 'is this A/B test significant', 'did the variant actually win', 'compute lift and confidence interval', 'score a marketing experiment', 'run the promote-gate', or wants a statistically defensible promote/hold/reject call on an A/B or multivariate marketing experiment. Computes lift, a bootstrap confidence interval, and a significance test (Mann-Whitney U or two-proportion z) using pure Python standard library, without scipy or numpy. Every p-value is computed from the input data, never estimated. Use before promoting any experiment variant to production.

This skill should be used when the user asks to draft a privacy policy, terms of service, cookie policy, or data processing agreement; when they ask about GDPR, CCPA, HIPAA, or other privacy regulations; when they need a compliance audit, legal gap analysis, or regulatory guidance; when they ask about employment law, IP rights, open source licensing, or contract review; when they mention 'legal', 'compliance', 'regulation', 'liability', 'terms', 'privacy', or 'lawsuit'. This skill also applies to crypto and digital asset questions: token classification (Howey test), security token offerings, stablecoins, GENIUS Act, DeFi compliance, CFTC jurisdiction, DAO liability, IRS crypto tax, AML/FinCEN MSB registration, tokenization of real-world assets, UCC Article 8, and smart contract legal review. Also use for designing or building agentic legal workflows, multi-agent compliance pipelines, and legal AI system architecture using Vercel AI SDK or CrewAI.

Explicit-only installer for the Product Skills Anthony and Caal Codex custom agents. Use ONLY when the user explicitly asks to install, update, check, uninstall, or set up Product Skills agents in Codex, including "install Anthony in Codex", "install Caal in Codex", "update the Product Skills Codex agents", or "check product_skills_legal". Never auto-invoke for ordinary legal, compliance, SOC 2, SEO, marketing, CRO, copywriting, or launch work.

This skill should be used when the user asks to "collect SOC 2 evidence", "build an evidence register", "prepare evidence for the auditor", "what artifacts do we need", "organize our control evidence", "respond to an auditor request list", or mentions evidence gathering, control artifacts, audit requests, screenshots, exports, or testing records for SOC 2.

This skill should be used when the user asks to "prepare for SOC 2", "run a SOC 2 gap analysis", "check our audit readiness", "map our controls", "what controls are missing", "review us for SOC 2 Type I", "review us for SOC 2 Type II", or mentions SOC 2, trust service criteria, control gaps, auditor prep, trust center readiness, or remediation planning.

This skill should be used when the user asks to "draft a SOC 2 policy", "write an access control policy", "write an incident response policy", "create security policies for audit", "prepare policy documents for SOC 2", "draft our control narratives", or mentions policy drafting, approval cadence, review frequency, governance language, or auditor-facing policy documentation for SOC 2.