Hektor — AI Trust & Safety Engineer AI Skill
Instant download · 30-day money-back guarantee. Pay once, keep forever — no subscription. Refund policy
Defend the platform: write enforceable policy, tune thresholds on real traffic, and keep appeals and reviewers healthy.
- Policy written so a classifier and a reviewer both apply it
- Threshold selection on real traffic precision and recall
- Over-enforcement measured as carefully as under-enforcement
- Review queues, reviewer wellbeing, appeals and transparency reporting
Trust and safety teams on AI products who need enforcement that survives both abuse and scrutiny.A trust and safety engineer commands $145+/hr, this is one file, yours forever.
Drop Hektor into Claude and get a trust and safety engineer who measures over-enforcement as carefully as under-enforcement, because a blunt countermeasure destroys legitimate users.
Hektor runs platform-scale abuse and harmful-content defence for AI products: policy written so a classifier and a human reviewer can both apply it; classifier development and threshold selection against precision and recall on real traffic; human review workflows, queue design and reviewer wellbeing; appeals and reversal processes; measuring both over-enforcement and under-enforcement; adversarial and coordinated abuse patterns; age assurance and minor safety; procedural escalation paths for illegal content; regional legal differences in what must be removed; transparency reporting; incident response for a safety failure; and red-team findings fed back into policy. This is defensive platform integrity work only.
What you get
- →Policy written so a classifier and a reviewer both apply it
- →Threshold selection on real traffic precision and recall
- →Over-enforcement measured as carefully as under-enforcement
- →Review queues, reviewer wellbeing, appeals and transparency reporting
How to install
Download the .skill package → open Claude → paste SKILL.md into your Project Instructions or system prompt → describe your requirement → Hektor builds the answer. Includes a full worked example so you see exactly what you get.
# Hektor - AI Trust & Safety Engineer You are Hektor, an AI trust and safety engineer. Every enforcement decision has two error types and you measure both. ## How you work 1. Write policy a classifier and a human reviewer can both apply 2. Set thresholds on real traffic, reporting precision and recall 3. Measure over-enforcement alongside under-enforcement, always 4. Design queues, appeals and reviewer wellbeing as first-class parts Never ship a countermeasure without measuring what it costs legitimate users, and escalate illegal content through the defined procedure only.
Excerpt from the actual file you'll download.
Four steps. Any AI chat.
- 01Download the file
After checkout, the download link lands in your inbox. Save the file anywhere on your device.
- 02Open your AI chat
Claude, ChatGPT, Gemini, Grok, or Copilot — whichever one you already use.
- 03Paste the file contents
Drop it into the system prompt, Project instructions, or custom instructions field.
- 04Start working
Your AI is now configured as a specialist. Ask it anything inside its domain.
No technical knowledge required. No subscription. Pay once, keep forever.
Works with every major AI chat.
Drop the file into your AI's system prompt, Project instructions, or custom instructions. No setup. No code. No vendor lock-in.
- Claude
- ChatGPT
- Gemini
- Grok
- Copilot
Works with any AI chat that accepts a system prompt or custom instructions.
Ready to specialise your AI?
One drop-in file. Pay once, keep forever — works with Claude & ChatGPT.