More than 3,400 systems are affected by malware that mines cryptocurrency and looks for new systems to infect, Lumen reports.
Its operators use a poem to change where i…
@hacks.gr
@hacks.grMore than 3,400 systems are affected by malware that mines cryptocurrency and looks for new systems to infect, Lumen reports.
Its operators use a poem to change where i…
@hacks.grMore than 3,400 internet-facing systems are reportedly infected in the Canto Incognito campaign.
Attackers use their computing power to mine cryptocurrency; some a…
https://en.hacks.gr/ereynites-pano-apo-3-400-systimata-molynthikan-gia-na-paragoyn-kryptonomismata/
@setevoy.bsky.socialLiteLLM: PostgreSQL Failed Requests, AWS RDS, and “Manual Debugging”
https://rtfm.co.ua/en/litellm-postgresql-failed-requests-aws-rds-and-%e2%80%9cmanual-debugging%e2%80%9d/
This one turned out to be a bi
#AWSRDS #LiteLLM #PostgreSQL
@wideareaai.bsky.socialThis shim accepts the Anthropic-format requests and re-emits them as OpenAI calls that your local server can actually understand. #selfhosting #batchinference #llm #localai #litellm #aiinfrastructure 2/2
@hasamba72.bsky.socialUnpatched account takeover in LiteLLM (CVE-2026-93355). JWT auth falls back to unverified email claims, allowing login as proxy_admin and permanent SSO rebinding. CVSS 8.8. #LiteLLM #CVE #AppSec https://www.ox.security/blog/litellm-an-ordinary-login-token-can-become-someone-elses-admin-account/
@hasamba72.bsky.socialLiteLLM Lens automates AI agent trace analysis in production. It investigates failures and groups recurring problems based on user-defined questions using OpenTelemetry and ClickHouse. #litellm #ai_agents #tool https://docs.litellm.ai/docs/proxy/lens
@hookflowai.bsky.social@setevoy.bsky.socialAI: LLM Spend Control with LiteLLM Budgets and OpenAI Limits
https://rtfm.co.ua/en/ai-llm-spend-control-with-litellm-budgets-and-openai-limits/
The main goal we had when intr
#AI #Grafana #LiteLLM
@diesec.bsky.socialCVE-2026-59822 (CVSS 8.8): LiteLLM MCP proxy lets attackers skip OAuth2 login entirely, no credentials needed. First MCP flaw in CISA KEV. Exploited since Sep 1. Upgrade to v1.84.0+. Federal deadline: today.
diesec.com/2026/09/lite...
diesec.com/de/2026/09/l...
#cybersecurity #diesec #LiteLLM
@setevoy.bsky.socialLiteLLM: Debugging AI Cost Monitoring with VictoriaMetrics
https://rtfm.co.ua/en/litellm-debugging-ai-cost-monitoring-with-victoriametrics/
Had a pretty interesting case
#AI #LiteLLM #LLM
@setevoy.bsky.socialLiteLLM: Custom Callbacks and LLM Evaluations with Judge LLM
https://rtfm.co.ua/en/litellm-custom-callbacks-and-llm-evaluations-with-judge-llm/
A quick recap of what we are d
#AI #LiteLLM #LLM
@stackflag.bsky.socialCVE-2026-37004 - litellm
The litellm library used by BerriAI, in versions up to 1.82.4, can be tricked into running any operating‑system command if someone sends a specially crafted request to its /prompts/test…
Too many irrelevant or confusing CVEs? Use stackflag.com
@thedailytechfeed.comLiteLLM’s default keys and guardrail bugs let attackers run root code and steal cloud credentials. #Security #AI #AIgateways #LiteLLM #CloudSecurity #Vulnerability https://thedailytechfeed.com/litellm-flaw-lets-attackers-run-root-code-steal-cloud-keys/
@setevoy.bsky.socialLiteLLM: Custom Callback for Traffic Mirroring and OTel Tracing to VictoriaTraces
https://rtfm.co.ua/en/litellm-custom-callback-for-traffic-mirroring-and-otel-tracing-to-victoriatraces/
We're currently setting up our
#AI #LiteLLM #LLM
@nisrulz.comCaught my coding agent paying flagship prices for a typo fix. Most requests just don't need the big model. Turns out you can route cheap prompts to cheap models automatically.
Read more: crushingcode.nisrulz.com/blog/complex...
@hendryadrian.bsky.socialCISA added 7 exploited flaws to KEV, impacting SonicWall, Sangoma, JFrog, Starlette, Kestra, and LiteLLM. Attackers are using them for admin access, reverse shells, miners, and credential theft. #SonicWall #LiteLLM #XMRig