Researchers bypass 6 AI agent frameworks with a simple ‘Repeat-After-Me’ prompt
Researchers used a Repeat-After-Me prompt to make six AI agent frameworks run attacker-written tool calls.
66
Vendor
LangChain
- Mentions · 7d0
- Mentions · 30d0
- All mentions3
- First seenSep 17, 2026
- Last mentionedOct 12, 2026
Timeline
3 articles · newest first[AINews] not much happened today
Latent Space AI news digest covers Anthropic's Claude Code Projects, Google's managed agent APIs, TypeSafe's Jev classifier, and OpenAI's Astra for Law launch.
35
Loopjacking: Hijacking Human-in-the-Loop Approval
Loopjacking lets agent approvals authorize a different operation than the human reviewed.
64
Appears with
- Googlevendor · 2
- OpenAIvendor · 2
- Jevproduct · 1
- NVIDIAvendor · 1
- OpenAI Agents SDKproduct · 1
- LangGraph Agent Serverproduct · 1
- Model Context Protocolproduct · 1
- Agent Development Kitproduct · 1
- GPT-6 Astraai model · 1
- OpenAIorganization · 1
- Latent Spaceorganization · 1
- Dartmouth Collegeorganization · 1
- TypeSafevendor · 1
- Google DeepMindvendor · 1
- CrewAIvendor · 1
- Perplexityvendor · 1
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.