cd /news/ai-safety/opencode-flaw-enables-code-execution… · home › topics › ai-safety › article
[ARTICLE · art-155248] src=osintsights.com ↗ pub= topic=ai-safety verified=true sentiment=↓ negative

OpenCode Flaw Enables Code Execution via Malicious MCP Servers

A critical vulnerability rated 9.8/10 in OpenCode's MCP sign-in flow allows malicious MCP servers to execute code on users' machines, according to a disclosure that was privately reported on August 29 and made public on October 19. A fix shipped in OpenCode version 1.18.33 on September 28. The flaw matters because MCP servers are widely used to connect AI coding agents to external tools, so a malicious server can turn a routine sign-in into arbitrary code execution.

by read1 min views4 publishedOct 22, 2026

A critical flaw in OpenCode's MCP sign-in flow, rated 9.8/10 in severity, allows malicious MCP servers to execute code, posing a significant threat to users. This vulnerability was privately reported on August 29 and publicly disclosed on October 19, with a fix released in version 1.18.33 on September 28.

source & further reading
osintsights.com — original article LiteLLM Exposes GitHub Token Leak Behind Malicious Package Release Threat Actor Wields AI in Multiform Cybercrime Spree Microsoft Links ConnectWise Breach to China-Aligned Threat Group
── more in #ai-safety 4 stories · sorted by recency
── more on @opencode 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/opencode-flaw-enable…] indexed:0 read:1min 2026-10-22 · —