Small Agents, Smaller Blast Radius: Controlling What AI Can Do, (Fri, Oct 16th)
SANS argues defenders should build small least-privilege AI agents rather than caging general-purpose ones, citing Check Point's tool-call deviation detection.
45
- Mentions · 7d0
- Mentions · 30d3
- All mentions9
- First seenAug 25, 2026
- Last mentioned15d ago
Timeline
9 articles · newest firstToFu: Exploiting and Fuzzing the LLM-Orchestrator-Tool Trust Boundary
ToFu fuzzing found verified semantic bugs in 343 LLM tools across 118 MCP servers.
66
APEX-Harm: Evaluating Agent Safety in Shared Professional Workspaces
APEX-Harm finds script-injection attacks often succeed against frontier agents in shared workspaces.
64
LLM-Driven Multi-Agent Control for Skill-Based Smart Manufacturing
LLM agents controlling a simulated six-module factory reach a 93% solve rate and diagnose silent faults.
38
SkillSeek: Revisiting Agent Skill Retrieval at Marketplace Scale
SkillSeek shows standard retrieval matches LLM-mediated agent skill selection at about half the cost.
44
MCP was always a bad idea?
Simon Willison argues MCP still matters for controlled, auditable agent access to external services.
36
Your AI agents can now control your Google Home devices
Google launches early-access MCP server letting MCP-capable AI agents like Claude and ChatGPT control Google Home devices.
40
Inside 90 days of attacks on AI infrastructure
Wiz honeypots recorded 90 days of active campaigns attacking LiteLLM, MCP servers, and AI frameworks via RCE, blind prompt injection, and credential theft.
62
AI supply chain risk is showing up in developer workflows first
Zentera Systems CEO says AI supply chain attacks currently hit developer workflows first, advising segmentation over tooling and citing the Phantom Raven campaign.
30
Appears with
- ChatGPTai model · 2
- Claudeai model · 2
- Googlevendor · 2
- OPC UAproduct · 1
- Zentera Systemsorganization · 1
- Gemini 3.8 Flashai model · 1
- Meta Museproduct · 1
- Google Nestproduct · 1
- SANSorganization · 1
- Sonnet 5ai model · 1
- BGE-baseai model · 1
- Google Workspaceproduct · 1
- Google Homeproduct · 1
- SkillSeekproduct · 1
- Anthropicorganization · 1
- Google Cloudproduct · 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.