About referral-loop-review
A skill for designing and reviewing referral programs with attention to trust, incentives, attribution, and abuse risks. It produces a referral flow, incentive table, event schema, risk controls, and experiment plan.
Key features
- 01Map invitation and reward states
- 02Distinguish referral loop types
- 03Define incentives and attribution
- 04Identify fraud and abuse controls
- 05Create event schema and experiment plan
Use cases
- 01Design a referral flow from invite to reward
- 02Review incentive eligibility and reversal rules
- 03Plan referral tracking and experiments
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ sylphxai/ skills --skill referral-loop-reviewAfter installing, type/referral-loop-reviewin Claude Code to confirm it loaded.
Bundled filesagents/openai.yamlreferences/referral-loop-systems.md
Details
From the project READMEReferral Loop Review
Use this skill to make referrals a trusted growth loop rather than a spam incentive.
Workflow
- Identify inviter, invitee, reward, trigger moment, channel, attribution window, and fraud risk.
- Read
references/referral-loop-systems.md. - Map invitation, acceptance, qualification, reward pending, reward grant, cancellation/reversal, and support states.
- Separate product-led sharing, paid incentives, team collaboration, affiliate economics, and abuse controls.
- Produce loop design, incentive table, event schema, risk controls, and experiment plan.
When not to use
- Do not use for generic advice the base model already handles without this skill's specific artifact contract.
Guardrails
- Do not reward spam, misleading claims, fake scarcity, or coerced contact upload.
- Do not grant irreversible rewards before qualification and abuse checks.
- Make terms, eligibility, attribution, expiration, and reversal rules visible.
Output format
Referral context:
Loop type:
Referral flow:
- trigger, channel, invite, qualification, reward, reversal
Incentive design:
- inviter reward:
- invitee reward:
- eligibility:
- caps/controls:
Risks:
- <risk> -> prevention, detection, support response
Metrics:
- invite, accept, activate, qualify, reward, fraud, retention
Comments
Markdown supported