miscProhard
ProAgent
0ops_ctf
Task: AI agent service with MCP support and user-configurable MCP server URL. Solution: MCP Tool Poisoning via malicious tool descriptions that instruct the agent to call built-in read_file("/flag") and pass the contents back through a two-step workflow.
$ ls tags/ techniques/
ai_agentmcpmodel_context_protocoltool_poisoningindirect_prompt_injectionssh_tunnelreverse_tunnelllm_exploitationagent_abuseread_file
MCP Tool Poisoning via malicious tool descriptionsIndirect Prompt Injection through tool responsesSSH reverse tunnel for exposing local MCP serverTwo-step workflow exploitation
🔒
Permission denied (requires tier.pro)
Sign in to access full writeups
Sign in with GitHub to continue. No email required.
$sign in$ grep --similar
Similar writeups
- [misc][free]Prometheon— HackTheBox
- [web][Pro]Lab 205 — DockForge — SSRF in Webhook Test Endpoint— hackadvisor
- [web][Pro]Lab 345 — PrintForge — RCE via Ghostscript Command Injection— hackadvisor
- [web][Pro]Lab 224 — ModelVault — Path Traversal via OCI Manifest Digest— hackadvisor
- [infra][Pro]Kobold— hackthebox