Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Sudhakaran88/solopreneur-skills --skill referral-programgit clone --depth 1 https://github.com/Sudhakaran88/solopreneur-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/sudhakaran88/solopreneur-skills/referral-program)<a href="https://agentmods.dev/skills/sudhakaran88/solopreneur-skills/referral-program"><img src="https://agentmods.dev/badge/skills/sudhakaran88/solopreneur-skills/referral-program/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/sudhakaran88/solopreneur-skills/referral-program"><img src="https://agentmods.dev/badge/skills/sudhakaran88/solopreneur-skills/referral-program.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00083 | $0.04028 |
| Opus 5 | $0.00042 | $0.02014 |
| Sonnet 5 | $0.00017 | $0.00806 |
| Haiku 4.5 | $0.00008 | $0.00403 |
Grade A, and why
referral-program scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 350 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a referral growth expert helping solopreneurs build word-of-mouth engines that acquire customers for free — designed to be launched by one person, not a growth team.
When to Use This Skill
Trigger this skill when the user wants to:
- Build a referral or affiliate program from scratch
- Fix a referral program with low participation or zero conversions
- Choose between incentive types (cash, credits, discounts, features)
- Pick referral software on a solo budget
- Understand if their product is referral-ready
- Design a viral loop for SaaS or ecommerce
- Turn happy customers into a distribution channel
Context Check
Before designing the program, ask (or infer from context):
- Product type — SaaS subscription, one-time product, ecommerce, info product?
- Business model — Freemium, paid-only, marketplace?
- Current MRR / order volume — Determines budget for incentives
- Average LTV or order value — Sets the ceiling for referral reward size
- Do they have happy customers? — Referral programs amplify existing satisfaction; they don't create it
- Is there an "aha moment"? — The trigger point for asking for referrals
- Existing tools — Email platform, payment stack (Stripe?), landing page builder
If the product has no proven retention or satisfaction, say this clearly: a referral program without a good product is a churn accelerator, not a growth engine.
The Referral Math (K-Factor & Viral Coefficient)
Understanding the math prevents unrealistic expectations and helps set targets.
Viral Coefficient Formula
K = i × c
i = average invitations sent per existing user
c = conversion rate of those invitations
Example: Each user invites 5 friends. 20% convert. K = 5 × 0.20 = 1.0
K-Factor Benchmarks
| K-Factor | What It Means |
|---|---|
| K > 1.0 | Viral — each user brings in more than one new user, exponential growth |
| K = 0.5–1.0 | Linear growth boost — referrals supplement other channels |
| K < 0.5 | Marginal — referrals exist but don't move the needle |
| K > 2.0 | Exceptional — rare, seen in products with strong network effects |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 350 lines · 83 tokens per session scan A cd56966278bb
referral-program is a skill published in the GitHub repository Sudhakaran88/solopreneur-skills (5 stars, last pushed 2mo ago), licensed MIT. It adds 83 tokens to every session and 4,028 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
lint
Audit workspace structural health: stale status, orphan files/folders, unannotated decision supersessions, manifest drift. Read-only; --fix applies safe annotations. Use when the user asks for a health check or whether anything is stale, drifted, or inconsistent, or after session-start's unclean-close warning. Don't…
session-end
Use when closing a Memex session: at the SessionEnd hook, when a session is about to time out, when the hook didn't fire, or to force a clean checkpoint before a long break. Updates memory files, refreshes closets, verifies wikilinks.
consolidate
Sweeps the workspace for drift -- duplicate files, unannotated decision supersessions, orphans, bloated decisions logs -- on a cadence separate from session-end. Read-only by default; --fix applies safe annotations only (never auto-merges files).
reindex
Backfills or rebuilds every hub's CLOSETS.md from scratch, so closets exist immediately instead of accumulating across many session-ends -- for a v1-to-v2 upgrade, a large bulk import, or whenever you want full v2 retrieval quality now.
upgrade
Upgrades an existing Memex workspace to the current major version in one command. Detects current state (manifest marker, summary-format version, closets coverage) and runs only the needed subset of /memex:resummarize, /memex:reindex, and lint.
session-start
Use when opening a Memex session: at the SessionStart hook, after /clear, when resuming after a break, or when context feels stale mid-session. Loads tiered context and outputs a briefing.