Growth Model

Pack Availability Guard

Before telling the user to run a skill from another project-local pack, check .agents/project.json.enabled_packs. If the target pack is not enabled, recommend $pack install <pack> instead of the target skill. Global skills are always valid. Skills from this same pack are valid because the current skill is already running from that pack.

Growth Model — Reforge-Style Growth Loop Design

Invoke as $growth-model.

Report-First Approval Gate

Default to report-only: present findings, evidence coverage, assumptions, recommended artifact path, and proposed file changes in a pre-approval alignment page plus a concise conversation summary for user approval before creating or updating canonical research, spec, or task files.

Do not write or overwrite synthesized deliverables until the user explicitly approves, unless the user invoked an explicit write/update/fix mode or clearly asked to write files upfront. Raw evidence capture may be persisted before analysis when reproducibility requires it; report those raw paths separately and still gate synthesized research/report writes.

When stopping for approval, build and attempt to open the alignment preview page first, then ask the user to review it and approve, question, or request adjustments. Do not include Recommended next skill, Recommended next command, or downstream routing language. The approval request itself is the next action. Only emit next-skill routing after the approved artifact has been written or updated.

Designs compounding growth loops using Reforge methodology. Identifies and designs acquisition loops (viral, content, paid, sales), retention loops, and monetization loops that compound over time. Growth loops replace the traditional funnel model — each loop's output becomes its input, creating sustainable growth.

Default stance: assume the user has no insider knowledge of growth mechanics or channel performance. The growth model must stand on research, metrics evidence, and codebase reality before asking for user input. Ask for corrections, proprietary data, and hard constraints, not intuition.

Prerequisites

• Hard: research/metrics.md (or research/{app}/metrics.md) must exist. If not, tell the user to run $metrics first and stop.
• Hard: research/gtm.md (or research/{app}/gtm.md) must exist. If not, tell the user to run $gtm first and stop.
• Soft: Read these if they exist:
  • research/journey-map.md — retention stages, aha moment, habit loop
  • research/monetization.md — pricing model, revenue mechanics
  • research/hook-model.md — engagement patterns, trigger-action-reward-investment cycle

Process

0. App Scope Resolution (Monorepo Support)

Before checking prerequisites, determine the app scope:

1. If $ARGUMENTS specifies an app name matching a subdirectory of research/, use it.
2. If research/ contains subdirectories (excluding files), list them and ask the user which app to target. If the session is already in Plan mode and there are 2-3 concrete choices, prefer request_user_input; otherwise ask in plain text. If only one subdirectory exists, use it automatically.
3. If no subdirectories exist, proceed with flat structure (single-product mode).

When app scope {app} is active:

• Read/write research from research/{app}/ instead of research/

1. Load Context

• Read research/metrics.md (or research/{app}/metrics.md) — success targets, KPIs, activation/engagement/retention/growth metrics
• Read research/gtm.md (or research/{app}/gtm.md) — channels, acquisition strategy, pricing model, early traction tactics
• Read research/journey-map.md (or research/{app}/journey-map.md) if it exists — customer journey stages, aha moment, habit loop, churn triggers
• Read research/monetization.md (or research/{app}/monetization.md) if it exists — pricing tiers, upgrade triggers, revenue model
• Read research/hook-model.md (or research/{app}/hook-model.md) if it exists — engagement loops, trigger-action-reward-investment cycle
• Read CLAUDE.md, README, and key source files for product context

2. Research Growth Loops

Use WebSearch with 4-6 targeted queries:

1. Reforge methodology — "Reforge growth loops [category]"
2. Domain-specific models — "growth model [domain] examples"
3. Competitor growth strategies — "[competitor] growth strategy"
4. Viral mechanisms — "viral loop [product type]"
5. Content loops — "content loop B2B SaaS"
6. Compounding mechanisms — "compounding growth mechanisms [category]"

3. Identify Loop Candidates

Categorize potential loops from research and product context:

Acquisition loops — how new users are acquired:

• Viral — user invites user (referral, sharing, word-of-mouth)
• Content — content attracts user (SEO, UGC, community content)
• Paid — revenue funds acquisition (paid ads, sponsorships)
• Sales — revenue funds sales team (outbound, enterprise sales)

Retention loops — how users stay and deepen engagement:

• Engagement — usage deepens value (more data, personalization, history)
• Network — more users = more value (marketplace, social, collaboration)
• Switching cost — investment creates lock-in (integrations, customization, data)

Monetization loops — how revenue compounds:

• Expansion — usage drives upgrades (seat growth, tier upgrades, usage-based expansion)
• Cross-sell — one product leads to another (platform play, add-ons)

If the session is already in Plan mode and there are 2-3 concrete choices, prefer request_user_input; otherwise ask in plain text:

• "Here are the growth loop candidates I see for this product. Which feel realistic given your current stage, resources, and product type? Any I should add or remove?"

4. Design Primary Loop

For the highest-potential loop, design it in detail:

• Each step in the loop — what happens at each stage, who does what
• Conversion hypothesis between steps — what must be true for users to move from one step to the next (each is a hypothesis until validated)
• Compounding mechanism — what makes the output feed back as input, creating exponential rather than linear growth
• Key metric at each step — the measurable signal that this step is working
• Time-to-complete-cycle estimate — how long one full loop iteration takes

If the session is already in Plan mode and there are 2-3 concrete choices, prefer request_user_input; otherwise ask in plain text:

• "Is this primary loop realistic? What's the weakest step?"

5. Design Supporting Loops (1-2)

Design secondary loops that reinforce the primary:

• Show how they interact — which loop's output feeds another's input
• Identify where loops share steps or handoff points
• Keep these lighter than the primary — they support, not compete

6. Map Dependencies & Metrics

• Align each loop step with metrics from research/metrics.md
• Identify gaps where metrics don't cover loop steps
• Show loop-to-loop dependencies — where one loop's output feeds another's input
• Flag metrics that need to be added to track loop health

If the session is already in Plan mode and there are 2-3 concrete choices, prefer request_user_input; otherwise ask in plain text:

• "Does this growth model align with your metrics framework? Any metric gaps or misalignments?"

7. Populate Next Steps

Before writing, check which files exist to populate the ## Next Steps section contextually. Include a Recommended item (the single highest-impact next step given current project state) with a one-line reason, followed by Other options (2-4 alternatives). Use this format in the output:

Next Steps

Recommended: $spec-interview [top growth mechanism] — spec the highest-impact growth mechanism for implementation

Other options:

• IF specs/ exist and tasks/roadmap.md exists: $roadmap — Update roadmap with growth loop implementation work
• IF product is live or launching: $experiment [growth hypothesis] — Test the primary loop's weakest conversion hypothesis

Impact-aware adjustments:

• IF downstream impact is Major: prepend $reconcile-research — [N] conflicts found in downstream docs as the first item
• IF downstream impact is Minor: annotate relevant skill suggestions with "(stale — [brief description])"

8. Write Output

Only after the user confirms, write the output files.

9. Downstream Impact Check

After writing, check for downstream research documents that may be affected. Only check documents that exist on disk.

Downstream documents to check (use {app}/ prefix when app scope is active):

• research/metrics.md
• research/gtm.md

For each existing downstream document:

1. Read it — focus on sections that reference growth strategy, acquisition channels, retention mechanisms, or expansion metrics
2. Identify specific conflicts: claims, assumptions, or references that contradict what was just decided. Examples:
   • A growth metric that doesn't align with the loop steps defined here
   • Channel strategy assumptions that don't match the acquisition loops designed
   • Retention targets that don't account for the retention loop mechanics
3. Note each conflict: downstream file, section, the stale claim (quote it), and what it should now say

Classify the impact:

• None: No downstream documents exist, or no conflicts found. Skip display entirely.
• Minor (1-2 small conflicts): Display conflicts to user inline.
• Major (3+ conflicts OR a foundational assumption changed — e.g., primary acquisition loop contradicts GTM channel strategy, retention model fundamentally differs from metrics targets): Display conflicts and strongly recommend $reconcile-research.

Output

research/growth-model.md (or research/{app}/growth-model.md)

# Growth Model

> Based on: research/metrics.md, research/gtm.md[, research/journey-map.md, research/monetization.md, research/hook-model.md]
> Date: [current date]
> Methodology: Reforge Growth Loops

## Summary
[2-3 sentences: the growth thesis — what loops drive this product's growth, why they compound, and what stage they're appropriate for]

## Loop Inventory

| Loop | Type | Potential | Stage-Readiness |
|------|------|-----------|-----------------|
| [loop name] | Acquisition / Retention / Monetization | High / Medium / Low | [why it fits or doesn't fit current stage] |

## Primary Growth Loop: [Loop Name]

**Type**: [Acquisition / Retention / Monetization]
**Compounding mechanism**: [what makes the output feed back as input]
**Cycle time estimate**: [how long one full loop iteration takes]

### Loop Steps

1. **[Step name]**
   - What happens: [description]
   - Key metric: [measurable signal]
   - Conversion hypothesis: [what must be true to reach step 2]

2. **[Step name]**
   - What happens: [description]
   - Key metric: [measurable signal]
   - Conversion hypothesis: [what must be true to reach step 3]

3. **[Step name]**
   - What happens: [description]
   - Key metric: [measurable signal]
   - Conversion hypothesis: [what must be true to feed back into step 1]

### Why This Loop Compounds
[Explain the specific mechanism — why does more output create more input? What's the flywheel effect?]

### Weakest Step
[Which conversion hypothesis is most uncertain, and what would validate it]

## Supporting Loop 1: [Loop Name]

**Type**: [Acquisition / Retention / Monetization]
**Compounding mechanism**: [what makes the output feed back as input]
**Cycle time estimate**: [how long one full loop iteration takes]

### Loop Steps
[Same structure as primary, lighter detail]

### Interaction with Primary Loop
[How this loop's output feeds the primary loop's input, or vice versa]

## Supporting Loop 2: [Loop Name]

**Type**: [Acquisition / Retention / Monetization]
**Compounding mechanism**: [what makes the output feed back as input]
**Cycle time estimate**: [how long one full loop iteration takes]

### Loop Steps
[Same structure as primary, lighter detail]

### Interaction with Primary Loop
[How this loop's output feeds the primary loop's input, or vice versa]

## Loop Interaction Map

[How loops feed each other — which loop's output becomes another's input]

- **[Loop A]** step [N] output --> feeds **[Loop B]** step [M] input
- **[Loop B]** step [N] output --> feeds **[Loop A]** step [M] input

## Metrics Alignment

| Loop Step | Metric | Current Target | Gap? |
|-----------|--------|---------------|------|
| [Primary loop - step 1] | [metric from metrics.md] | [target] | [Yes — no metric covers this / No] |
| [Primary loop - step 2] | [metric from metrics.md] | [target] | [Yes / No] |
| [Supporting loop - step 1] | [metric from metrics.md] | [target] | [Yes / No] |

### Metric Gaps
[List loop steps that have no corresponding metric in research/metrics.md — these need to be added]

## Growth Hypotheses to Test

| Hypothesis | Loop | Priority | Validation Method |
|-----------|------|----------|-------------------|
| [conversion hypothesis between loop steps] | [which loop] | High / Medium / Low | [how to test — experiment, survey, instrumentation] |

## Strategic Implications

### What This Model Means for Product
[Feature priorities that support loop mechanics — what to build to strengthen the weakest steps]

### What This Model Means for GTM
[How growth loops should shape channel investment, messaging, and launch sequencing]

### What This Model Means for Metrics
[Which metrics need to be added or reframed to track loop health]

### Stage-Appropriate Sequencing
[Which loops to activate now vs. later — match loop complexity to current stage]

<!-- Only include when downstream impact is Minor or Major -->
## Downstream Impact

> Checked: [list of downstream docs checked]
> Impact: Minor | Major

### Conflicts Found

1. **research/[file].md** — [Section Name]
   - **Stale**: "[exact quote from downstream doc]"
   - **Now**: [what this skill's output says instead]

[For Major only:]
> **Recommended action**: Run `$reconcile-research` to audit and fix all affected downstream documents.

## Next Steps

**Recommended:** `$spec-interview [top growth mechanism]` — spec the highest-impact growth mechanism for implementation

Other options:
- [conditional items from step 7 — only include items whose conditions are met]

research/growth-model-search-log.md (or research/{app}/growth-model-search-log.md)

Raw research log — queries, findings, evidence for each growth loop decision.

Create the research/ directory if it doesn't exist.

Task Classification

When this skill produces follow-up work, file it by execution semantics:

• Immediately actionable implementation or documentation work goes in tasks/todo.md.
• Human-only external actions tied to automated steps go in tasks/manual-todo.md with _(blocks: Step N.X)_ or _(after: Step N.X)_; repo edits, SDK wiring, generated assets, local commands, tests, audits, and authenticated CLI/API work stays in tasks/todo.md.
• One-time condition-gated records, baselines, or future measurements go in tasks/record-todo.md with source, condition, non-blocking reason, evidence, and promotion rule.
• Cadence-based reviews, playtests, adoption checks, investor updates, retros, or docs-health checks go in tasks/recurring-todo.md with cadence, owner/agent, next due, evidence path, and escalation conditions.
• Do not put non-blocking records or recurring obligations in tasks/todo.md unless they have been explicitly promoted into current execution work.

Constraints

• Requires metrics + GTM. Growth loops without success targets and channel strategy are theoretical.
• Stage-appropriate. Don't design viral loops for pre-launch products with no users. Match loop complexity to current stage.
• Compounding is key. If the loop's output doesn't feed back as input, it's not a loop — it's a funnel step.
• Hypothesis-driven. Every conversion between loop steps is a hypothesis until validated.
• Present before writing. Never write output files until the growth model has been presented and validated.
• Do not overwrite existing research/growth-model.md (or research/{app}/growth-model.md) without asking the user first.

Alignment Page

When this skill produces durable deliverables (research, specs, plans, reports, prototypes, or any document output), build a full-depth HTML alignment page at alignment/growth-model-{topic}.html. Use a normalized topic slug derived from the app, feature, research subject, report subject, or output filename.

Full content requirement. The alignment page must contain the complete content of every proposed markdown deliverable -- every section, every finding, every detail, every list item. It is a thorough interactive review document, not a summary. Render the full deliverable content in clean, readable HTML with appropriate hierarchy, styling, and navigation. If the skill writes multiple scoped deliverables in one run, build one alignment page that contains all deliverables with anchor-linked navigation. Durable tracker artifacts, such as research/assumption-tracker.md, remain canonical markdown outputs but must also be fully rendered into the alignment page before approval.

Dark-mode styling. Use a dark color scheme by default. Base CSS variables: --bg: #0d1117; --surface: #161b22; --border: #30363d; --text: #c9d1d9; --text-muted: #8b949e; --accent: #58a6ff; --green: #3fb950; --red: #f85149; --orange: #d29922; --purple: #bc8cff;. Apply background: var(--bg); color: var(--text); on body. Use --surface for cards, nav, and table headers. Use --border for all borders. Use --purple for question blocks and gate headings. Use --accent for links and section headings. Keep headings color: #fff or var(--accent) for hierarchy. Question block backgrounds should use #1c2333.

Alignment gates. Treat gates as explicit review sections inside the HTML page. A gate blocks finalization until its required inline questions are answered and compiled into YAML. Include every gate that applies to the skill output, and include these gate types whenever relevant: evidence coverage, assumptions/confidence, scope/non-goals, candidate/verdict decisions, artifact destination, proposed file changes, coverage checkpoint, and post-approval route.

Report-only research gates. For report-only or pre-approval research skills, the alignment page must explicitly contain evidence coverage, assumptions/confidence, recommended path, proposed file changes, and approval gates before any canonical research, spec, or task file is created or updated.

Required inline questions. Each gate must contain at least one required inline question placed directly under the content it governs, inside a visually distinct question block. Each question must use radio-button inputs and include two standing options after the skill-generated choices: "Other / None of the above" backed by a multi-line text box for free-form input, and "Need clarification" backed by an optional notes box where the user can explain what is unclear. When any radio option other than "Other" or "Need clarification" is selected, show an optional "Additional notes" text box beneath it so the user can qualify their choice. Generate questions based on what genuinely needs user input -- do not add filler questions. Do not create a separate bottom "Decisions & Clarifications" section.

Gate YAML contract. At the bottom of the page, include a "Compile Answers" button that aggregates answers from all inline gate questions throughout the page, including free-text notes. The button remains disabled until every required question has a selection, shows a count of remaining unanswered questions, and scrolls to the first unanswered question if clicked early. When every question is answered, generate a structured YAML block with one item per gate answer using this stable shape: section, gate_type, status (answered, other, or needs-clarification), answer, optional notes, and optional target_artifact or target_path when the gate controls file output. After successful compilation, automatically attempt to copy the YAML to the clipboard with the Clipboard API, display copy status, and display the YAML in a read-only textarea with an explicit "Copy YAML" button. The copy button must retry clipboard copy when supported and fall back to selecting the textarea contents when clipboard access is unavailable or blocked.

Pre-approval stop. Before user approval, the next action is review of the HTML alignment page, not downstream routing. Ask the user to review the page and provide the compiled YAML answers. Do not include Recommended next skill, Recommended next command, or downstream routing language until after compiled YAML has been provided and the approved artifacts have been written or updated.

Diff highlighting on updates. When the agent updates an existing alignment page after receiving compiled answers, highlight what changed since the previous version. The agent chooses inline annotation or side-by-side layout per situation.

Archiving. Before replacing an existing alignment page, archive it to docs/history/archive/YYYY-MM-DD/HHMMSS/alignment/growth-model-{topic}.html.

Browser open. Attempt to open the resulting HTML page in the browser and report whether the open succeeded or was blocked. A blocked browser-open attempt does not make the skill fail when the files were written correctly.

Archive-First Replacement Policy

• Before replacing or substantively rewriting an existing canonical research/spec document (research/**/*.md, specs/**/*.md, or docs/specifications/**/*.md), copy the current file to docs/history/archive/YYYY-MM-DD/HHMMSS/<original-relative-path>.
• Preserve the archived snapshot exactly as it existed before the change; do not edit the archived copy after creating it.
• After the archive snapshot exists, write the updated document to the original canonical path.
• Report both the archive path and the updated canonical path in the final output.
• New files do not need archive snapshots. Append-only updates do not need archive snapshots unless an existing section is regenerated or rewritten.
• Keep any existing user approval requirement before overwriting or replacing a document; archiving does not replace asking when the skill already requires approval.

Default Shipping Contract

• Default next-step routing: when reporting completion, include either Recommended next skill: <command> or the two-line pair **Next work:** <specific task or "none"> and **Recommended next command:** <one command or route> so the next operator has a concrete handoff.
• If this skill creates or modifies tracked repository files, finish by committing and pushing all intended changes to the repository primary branch (main when present, otherwise master) before stopping, even if the user did not explicitly ask for commit/push.
• Do not leave tracked changes or unpushed commits behind. If unrelated tracked work is already present, either include it in sensible commits too or stop and explain the blocker.
• This contract does not override stricter safety rules about secrets, destructive history changes, release publication/tag confirmation, or production deploy confirmation.