feat: add product growth audit workflow

README.md

@@ -17,6 +17,7 @@
 | [Deep Project Audit](prompts/project-audit.md) | You need to understand an unfamiliar or risky codebase | Architecture map, operational picture, prioritized findings, and a remediation plan |
 | [Idea Discovery Interview](prompts/idea-discovery-interview.md) | You have only a rough product or feature idea | A recorded discovery brief and an evidence-based implementation plan |
 | [Security Red-Team Audit](prompts/security-red-team-audit.md) | You are authorized to security-test a project | Threat model, verified findings, safe remediation, and regression tests |
+| [Product Growth Audit](prompts/product-growth-audit.md) | An existing app needs a clearer improvement direction | Evidence-based UI/UX, product, business, and technical roadmap |
 
 ```text
 Capture context  →  Understand the system  →  Build the right thing

plugins/agent-workflows/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "agent-workflows",
-  "version": "0.2.0",
+  "version": "0.3.0",
   "description": "Portable Markdown workflows for AI coding agents.",
   "author": {
     "name": "bglglzd",
@@ -9,7 +9,7 @@
   "homepage": "https://github.com/bglglzd/agent-workflows",
   "repository": "https://github.com/bglglzd/agent-workflows",
   "license": "MIT",
-  "keywords": ["ai-agents", "prompts", "skills", "codex", "developer-tools", "security"],
+  "keywords": ["ai-agents", "prompts", "skills", "codex", "developer-tools", "security", "product-management", "ux"],
   "skills": "./skills/",
   "interface": {
     "displayName": "Agent Workflows",
@@ -22,7 +22,8 @@
       "Use $session-handoff to preserve this project's context.",
       "Use $project-audit to perform a deep, read-only audit of this project.",
       "Use $idea-discovery-interview to turn my idea into a buildable plan.",
-      "Use $security-red-team-audit to assess this authorized project and fix verified findings."
+      "Use $security-red-team-audit to assess this authorized project and fix verified findings.",
+      "Use $product-growth-audit to identify and plan the highest-leverage improvements for this app."
     ],
     "brandColor": "#0EA5E9"
   }

plugins/agent-workflows/skills/product-growth-audit/SKILL.md

@@ -0,0 +1,27 @@
+---
+name: product-growth-audit
+description: Audit an existing application to discover and prioritize its best product, UI/UX, business, growth, reliability, and technical improvements. Use when Codex must turn an unclear improvement backlog into an evidence-based opportunity map and 30/60/90-day plan without assuming a mass-market strategy or inventing metrics.
+---
+
+# Product Growth Audit
+
+Find the strongest next improvements for an existing product. Produce a roadmap based on evidence, not a generic feature list.
+
+## Workflow
+
+1. Inspect the repository, docs, current UI, tests, issues, release notes, feedback, and data that are actually available. Record verified facts, assumptions, and unknowns.
+2. Map users, jobs to be done, product promise, key workflows, platform, maturity, constraints, and the journey from first touch through repeat use and support.
+3. Evaluate the relevant lenses: positioning; UX/UI; core value and features; activation, retention, distribution, and business model; trust and reliability; technical leverage; and AI quality when AI is part of the product.
+4. Evaluate niche, internal, B2B, open-source, and mass-market paths from the available evidence. Do not assume broad distribution is the goal.
+5. Create a non-duplicative opportunity map. For every item, provide evidence, user problem, improvement hypothesis, expected outcome, success signal, impact/confidence/effort rationale, dependencies, risks, and a smallest safe validation step.
+6. Separate quick quality wins, strategic bets, growth or business options, and technical enablers.
+7. Write or update product documentation. When no convention exists, use `docs/product/PRODUCT_GROWTH_AUDIT.md`.
+8. Deliver a 30/60/90-day roadmap split into Now, Next, and Later. Include scope, acceptance criteria, measurement plan, dependencies, and an owner type for every recommended item.
+
+## Guardrails
+
+- Do not invent metrics, users, competitor claims, market demand, revenue, or evidence. Mark missing data clearly.
+- Do not recommend trendy features, AI, gamification, paywalls, or growth hacks without a product-specific rationale.
+- Use external research only with authorization and cite it.
+- Do not implement, redesign, alter production, publish marketing, or contact users before approval.
+- Preserve real differentiators and prioritize feasible work for the actual team, codebase, budget, and privacy posture.

plugins/agent-workflows/skills/product-growth-audit/agents/openai.yaml

@@ -0,0 +1,4 @@
+interface:
+  display_name: "Product Growth Audit"
+  short_description: "Prioritize an app improvement roadmap"
+  default_prompt: "Use $product-growth-audit to identify high-leverage improvements for this app."

prompts/product-growth-audit.md

@@ -0,0 +1,78 @@
+# Product Growth Audit & Improvement Plan
+
+Use this when an application already exists but its next best improvements are unclear. It works for consumer, B2B, internal, niche, and open-source products — not only mass-market apps.
+
+## Prompt
+
+Perform a product growth audit of this existing application. Combine product strategy, UX/UI critique, customer-value analysis, business thinking, and technical judgment to identify the highest-leverage improvements. Produce an evidence-based roadmap, not a generic feature brainstorm. Do not implement, redesign, publish marketing, contact users, or change production systems until I approve a chosen plan.
+
+### Ground rules
+
+- Start with evidence from the repository, product documentation, current UI, tests, issues, support feedback, analytics, release notes, and any data I explicitly provide.
+- Do not invent users, metrics, revenue, competitors, market demand, or analytics. Mark missing evidence as an assumption or a research question.
+- Do not assume the product should serve a broad market. Evaluate whether its best path is a focused niche, internal workflow, prosumer tool, B2B product, open-source project, or something else.
+- Preserve the product's real differentiators. Do not recommend AI, social features, gamification, paywalls, or growth tactics merely because they are fashionable.
+- Use external market or competitor research only when I authorize it; cite sources and separate observed facts from inferences.
+- Keep recommendations feasible for the actual team, codebase, budget, platform, privacy posture, and operating environment.
+
+### Understand the product first
+
+1. Inspect the repository and available product materials. Identify who the product appears to serve, its promise, key workflows, platform, current capabilities, technical constraints, business model if any, and current maturity.
+2. Map the user journey from discovery or installation through activation, first value, repeated use, support, upgrade, and exit. Include onboarding, empty states, error states, settings, permissions, and recovery paths.
+3. Build a short product snapshot with **verified facts**, **assumptions**, **unknowns**, and the highest-value questions that would change the roadmap.
+4. If a safe local build or demo exists, experience the primary flows as a user. Do not use production accounts or alter real data without explicit authorization.
+
+### Audit every improvement surface
+
+Evaluate the product through these lenses, using only the ones that apply:
+
+1. **Problem, user, and positioning** — job to be done, user segments, urgency, alternatives, differentiation, and clarity of the product promise.
+2. **UX and UI** — information architecture, navigation, onboarding, visual hierarchy, interaction feedback, copy, defaults, empty/loading/error states, responsive behavior, keyboard flow, accessibility, localization, and perceived quality.
+3. **Core value and features** — workflow friction, missing capabilities, integrations, automation, collaboration, import/export, configurability, trust, and the smallest improvement that unlocks more value.
+4. **Activation, retention, and distribution** — time to first value, reasons to return, documentation, templates, demos, sharing or referral loops, developer adoption, sales-assisted motion, community, app-store or landing-page needs, and niche-specific distribution. Consider monetization only when relevant.
+5. **Trust, reliability, and quality** — privacy, security, permissions, data ownership, backups, performance, offline behavior, error recovery, supportability, observability, accessibility, and release confidence.
+6. **Technical leverage** — architecture bottlenecks, tests, design system, API quality, cost, maintainability, instrumentation, operational toil, and improvements that make future product work faster or safer.
+7. **AI-specific product quality** — only if AI is part of the product: control, transparency, latency, cost, evaluation, failure recovery, user trust, data boundaries, and whether AI solves a real user problem.
+
+### Turn observations into opportunities
+
+Create a non-duplicative opportunity map. Seek breadth, but prefer a smaller set of strong opportunities over an unprioritized idea dump. For every opportunity, include:
+
+- category and affected user or workflow;
+- evidence and current pain or missed value;
+- the improvement hypothesis and proposed change;
+- expected user outcome and business or product outcome when relevant;
+- success signal or metric to measure;
+- impact, confidence, and effort as **High / Medium / Low**, with rationale;
+- dependencies, risks, and the smallest safe experiment or validation step.
+
+Separate quick quality wins, strategic product bets, business or distribution bets, and technical enablers. Do not score an opportunity as high-confidence without evidence.
+
+### Required deliverable
+
+Write or update the project's established product documentation. If none exists, create `docs/product/PRODUCT_GROWTH_AUDIT.md` with this structure:
+
+```markdown
+# Product growth audit
+
+## Product snapshot: facts, assumptions, and unknowns
+## Primary users, jobs, and current value proposition
+## Journey and experience map
+## Findings by lens
+## Opportunity map
+| Priority | Opportunity | Evidence | Expected outcome | Impact | Confidence | Effort | Validation |
+## Recommended product direction
+## UI/UX improvements
+## Technical enablers and quality work
+## Growth, business, and distribution options
+## 30 / 60 / 90-day roadmap
+## Questions that need answers before commitment
+```
+
+Prioritize the roadmap into:
+
+- **Now** — the next one or two validated, high-leverage actions;
+- **Next** — work that follows after the first outcomes are measured;
+- **Later** — larger bets, dependencies, or ideas that need more evidence.
+
+For each roadmap item, specify an owner type, scope, expected outcome, acceptance criteria, measurement plan, and dependency. End with the top three recommended next actions and ask for approval before implementation.