←
OffByQuant22 posts
OffByQuant avatar

OffByQuant

@OffByQuant

github.com/OffByQuant

◷ Joined June 2026▧ 2 Photos and videos
OffByQuant avatar
OffByQuant@OffByQuant·Nov 16

open-appsec uses machine learning to flag suspicious web and API traffic. Its bundled Basic model is recommended for monitoring and testing. Review Detect/Learn alerts before considering blocking. github.com/openappsec/openap… #AppSec #APISecurity #WAF #AI

OffByQuant avatar
OffByQuant@OffByQuant·Nov 12

#NetSentinel v1.16.0 adds UniFi IPS alerts to a Windows monitor with local ML anomaly detection. Use outliers as review leads. Live capture needs separately licensed Npcap. github.com/jamesccupps/NetSe… #NetworkSecurity #ThreatDetection

OffByQuant avatar
OffByQuant@OffByQuant·Nov 12

#STRIDE-GPT v0.21.0 adds stable threat IDs and requires source evidence for cross-component threats. Use matched excerpts to review model-generated hypotheses, then validate the risk. github.com/mrwadams/stride-g… #ThreatModeling #AppSec

OffByQuant avatar
OffByQuant@OffByQuant·Nov 10

#OWASP #DockSec uses a local model to explain Dockerfile and Compose scan findings, including risks that may interact. Check the scanner evidence before accepting a suggested fix. github.com/OWASP/DockSec #ContainerSecurity #DevSecOps

OffByQuant avatar
OffByQuant@OffByQuant·Nov 10

Newly discovered #TriageWall puts #Suricata and #Wazuh alerts in one review queue. A prefilter handles repeat noise, then a local model triages the remaining alerts. Check the evidence before closing an alert. github.com/aaronphifer/triag… #SecOps #NetworkSecurity

OffByQuant avatar
OffByQuant@OffByQuant·Nov 6

#Shannon Lite uses source code to guide active web/API #security tests. Probes can change target state. Use authorized disposable staging and verify findings. github.com/keygraphhq/shanno… #AppSec #DAST

OffByQuant avatar
OffByQuant@OffByQuant·Nov 6

#Prowler's Nov 5 source update adds #AWS #Bedrock #AgentCore checks for #VPC mode, payment-role separation and #CloudTrail data events. Use the findings to guide posture review. The checks are in merged source. github.com/prowler-cloud/pro… #AISecurity #CloudSecurity

OffByQuant avatar
OffByQuant@OffByQuant·Nov 3

#qwen-3.8-flash-next vs #glm-5.3, running on an #RTX5090. GLM performed far better in this real-world test than its benchmark score suggested. It delivered a stunning working page first try; Qwen needed a repair round.

OffByQuant avatar
OffByQuant@OffByQuant·Nov 2

Newly discovered: established #AiSOC uses local AI for alert triage. v21 adds model provenance and closure accounting. Effectiveness unverified; review policy for automated closures. github.com/beenuar/aisoc #SOC #AISecurity

OffByQuant avatar
OffByQuant@OffByQuant·Oct 31

Geiger inventories local #AI agents and #MCP configuration, with baseline comparisons for drift. No model or #API account needed. It reads configuration, not runtime behavior. github.com/Atomburstofficial… #AISecurity #MCPSecurity

OffByQuant avatar
OffByQuant@OffByQuant·Oct 31

Review #AI-agent skills before installing them. SkillSpector checks for dangerous code, prompt-injection patterns and #MCP risks. Static scans need no paid model #API; findings still need review. github.com/NVIDIA/SkillSpect… #AISecurity #OpenSource