Macawi
@macawiai.bsky.social
Pioneering convergent AI cybersecurity. Zero Trust Agentic Security. Threats evolve - We evolve faster. https://macawi.ai
We built Strigoi (open source, think Metasploit for agentic AI) so you can scan your environment NOW.
Free security scans available: jamie.saker@macawi.ai
Because "I told you so" doesn't help if you're already breached.
github.com/macawi-ai/st...
[4/4]
Free security scans available: jamie.saker@macawi.ai
Because "I told you so" doesn't help if you're already breached.
github.com/macawi-ai/st...
[4/4]
GitHub - macawi-ai/Strigoi: Interactive CLI for AI/LLM security assessment. Directional reconnaissance framework for modern AI infrastructure testing.
Interactive CLI for AI/LLM security assessment. Directional reconnaissance framework for modern AI infrastructure testing. - macawi-ai/Strigoi
github.com
January 11, 2026 at 10:46 PM
We built Strigoi (open source, think Metasploit for agentic AI) so you can scan your environment NOW.
Free security scans available: jamie.saker@macawi.ai
Because "I told you so" doesn't help if you're already breached.
github.com/macawi-ai/st...
[4/4]
Free security scans available: jamie.saker@macawi.ai
Because "I told you so" doesn't help if you're already breached.
github.com/macawi-ai/st...
[4/4]
November 2025: Anthropic announces China exploited EXACTLY those MCP vulnerabilities we detailed.
Agentic AI code = new attack surface
Traditional security tools = blind to it
This isn't theory anymore. It's Tuesday.
www.anthropic.com/news/disrupt...
[3/4]
Agentic AI code = new attack surface
Traditional security tools = blind to it
This isn't theory anymore. It's Tuesday.
www.anthropic.com/news/disrupt...
[3/4]
Disrupting the first reported AI-orchestrated cyber espionage campaign
A report describing an a highly sophisticated AI-led cyberattack
www.anthropic.com
January 11, 2026 at 10:45 PM
November 2025: Anthropic announces China exploited EXACTLY those MCP vulnerabilities we detailed.
Agentic AI code = new attack surface
Traditional security tools = blind to it
This isn't theory anymore. It's Tuesday.
www.anthropic.com/news/disrupt...
[3/4]
Agentic AI code = new attack surface
Traditional security tools = blind to it
This isn't theory anymore. It's Tuesday.
www.anthropic.com/news/disrupt...
[3/4]
Back in Sept, we published this breakdown of Model Context Protocol vulnerabilities. Everyone building with Claude, AI agents, or MCP ignored it.
"That's theoretical" they said.
"Not a real attack vector" they said.
www.macawi.ai/research/blo...
[2/4]
"That's theoretical" they said.
"Not a real attack vector" they said.
www.macawi.ai/research/blo...
[2/4]
Macawi AI - Zero Trust Agentic Security
The First Zero-Trust Agentic Security Company
www.macawi.ai
January 11, 2026 at 10:45 PM
Back in Sept, we published this breakdown of Model Context Protocol vulnerabilities. Everyone building with Claude, AI agents, or MCP ignored it.
"That's theoretical" they said.
"Not a real attack vector" they said.
www.macawi.ai/research/blo...
[2/4]
"That's theoretical" they said.
"Not a real attack vector" they said.
www.macawi.ai/research/blo...
[2/4]
