Mohi@disismohi
Kali-GPT (commit a608c74) has a command injection in its MCP server. CVE-2026-117569. Prompt-injected agents can run arbitrary commands via unsanitized scanner parameters. This is the new supply chain.
1001071
94 followersView on X
CVE-2026-117569
LOWCVSS 8.5 · HIGHSignal is active with 2 mentions in latest observed window
Immediate actions
- Track advisory updates for patch or workaround availability
Recommended action window: Monitor and triage in normal cycle
NVD description
Kali-GPT through commit a608c74 contains an OS command injection vulnerability in execute_tool() in kali_gpt/tools/mcp_server.py that allows MCP clients to run commands via unsanitized scanner parameters. Attackers or prompt-injected agents can embed shell metacharacters in parameters like target, ports, domain, wordlist, or url, executed via subprocess.run with shell=True as the MCP server user.
0.0/ 10 priority
Sources & remediation
Other references
- https://github.com/alishahid74/kali-gpt
- https://github.com/alishahid74/kali-gpt/blob/a608c74e99910232dd208fee99de860458672b82/kali_gpt/tools/mcp_server.py#L229-L356
- https://github.com/alishahid74/kali-gpt/issues/12
- https://www.vulncheck.com/advisories/kali-gpt-through-commit-a608c74-os-command-injection-via-mcp-scanner-tool-parameters
Weakness type (CWE)
CWE-78
Priority
LOW
Exploitation
NONE
PoC
NONE
Patch
NONE
Momentum
NONE
Threat summary
- 2 mentions across 1 observed day
What's happening
- 2 total mentions across 1 day
Deep dive
Activity timeline2 mentions / 1d
Referenced assets1 URL
By indicator
Full discourse2 posts
- Mohi@disismohi
MCP adoption is accelerating. This CVE won't be the last. The pattern: LLM calls tool, tool trusts parameters, parameters came from a prompt injection three hops back. Defense: input validation at the server, not the client. source: https://nvd.nist.gov/vuln/detail/CVE-2026-117569
100004794 followersView on X