About abuse-fraud-risk-review
A skill for product teams reviewing abuse and fraud risks while accounting for harm to legitimate users. It structures the work into a risk matrix and control plan, with review and appeal paths.
Key features
- 01Identify assets, incentives, and abuse surfaces
- 02Assess harm to legitimate users
- 03Map signals, controls, and review paths
- 04Separate prevention, detection, response, and learning
- 05Produce a risk matrix and control plan
Use cases
- 01Review abuse risks in a product
- 02Plan fraud controls and escalation paths
- 03Assess false positives and recovery options
Install
Copy the command and run it in a terminal.
$
npx skills add https:/After installing, type/abuse-fraud-risk-reviewin Claude Code to confirm it loaded.
Bundled filesagents/openai.yamlreferences/abuse-fraud-risk-patterns.md
Details
From the project READMEAbuse Fraud Risk Review
Use this skill to protect product integrity without punishing legitimate users unfairly.
Workflow
- Identify assets at risk, attacker incentives, abuse surfaces, and legitimate-user harm.
- Read
references/abuse-fraud-risk-patterns.md. - Map signals, controls, review ladder, support route, and recovery/appeal.
- Separate prevention, detection, response, and learning.
- Produce risk matrix and control 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 recommend silent irreversible punishment without evidence and appeal path.
- Do not collect excessive data for vague fraud goals.
- Balance false positives against abuse damage.
Output format
Abuse surface:
Assets at risk:
Risk matrix:
- <scenario> -> likelihood, impact, controls, review path
Controls:
- Prevent:
- Detect:
- Respond:
- Recover:
Events/signals:
- <signal>
Comments
Markdown supported