# Abuse and trust-and-safety review agent

For platforms that host user content, reported URLs are analysed visually and textually for phishing, spam or copyright issues; the agent proposes an action and a security engineer makes the final call.

**Why it matters:** False positives cause wrongful takedowns; misses leave harm online. The agent does the formulaic first pass.

- Trigger: Abuse report in queue
- Tools: Vercel AI SDK, Browserless, Claude, OpenAI
- Time saved (rough estimate): 5–15 hours/week
- Typical agency build cost (rough estimate): $8k–$25k

## Steps
1. Fetch and screenshot URL
2. Visual + text analysis
3. Recommended action + evidence
4. Engineer decides
5. Record decision, close ticket

## Pitfalls
- Keep a human on every enforcement action.
- Log evidence for appeals.

## Sources
- [Vercel: what we learned building agents at Vercel](https://vercel.com/blog/what-we-learned-building-agents-at-vercel)

---
Source: https://cofounderbase.com/workflows/trust-safety-review-agent/?utm_source=md&utm_medium=ai · Cofounderbase, a directory of agencies that build AI internal tools and automations. Data comes from each agency's own website and official partner directories; rankings are not sold.
