Agentjacking
Named by 4 kept items in this workspace.
Search every workspace for Agentjacking. It matches the name as written, so it can find other things that share it.
The items that name it 4
Attacker POSTs fabricated Sentry error events (using a public DSN) formatted to mimic the Sentry MCP server's own output, embedding an npx command in a '## Resolution' section that the agent executes when asked to fix issues; 85% success rate across Claude Code, Cursor and Codex; 2,388 vulnerable or
Attack in which a single fake bug report with a poisoned log turns a routine triage request into remote code execution in an AI coding agent. It needs no malware or exploit and triggered no security control. Tenet Security confirmed it in the wild across 2,388 organizations, including 71 of the top
Attack technique targeting agentic platforms discussed alongside Mastra compromise.
Attackers inject malicious instructions into Sentry error events; AI coding agents retrieve these via MCP and execute the instructions with the developer's elevated permissions. Described by hosts as sophisticated prompt injection stemming from an MCP server failing to sanitize third-party data befo
All techniques in this workspace.