MCP · TOOL POISONING
MCP Server Pentesting
Penetration testing of Model Context Protocol servers and agent tool integrations before they get production credentials.
5
Deliverables
5
Tools
2
Attacks covered
What I deliver
- Tool description review for poisoning
- Scope and permission review per tool
- Rug-pull and version-change detection
- Tool shadowing across multiple servers
- Data-flow review: untrusted input to outbound tools
⟩ how I approach it
Step by step
- 01
List every connected MCP server and tool
- 02
Read tool descriptions for poisoning
- 03
Check scopes and credentials
- 04
Test tool shadowing and rug-pulls
- 05
Pin, hash and monitor definitions
⟩ who it's for
Teams using AI agents with MCPCompanies connecting LLMs to internal dataDevelopers publishing MCP servers
⟩ real-world projects
Selected engagements
From my work at BERRY9 IT Services, as a freelancer, at CRIS InfoTech and Hacktr Technologies, and my own tools and research.
01
MCP server reviews at CRIS InfoTech — tool descriptions, scopes and data flows checked before production use
⟩ common vulnerabilities & how I find them
Top bugs I hunt — and the steps
The recurring weaknesses I see in this domain, with the repeatable workflow I run from discovery to fix.
- STEP 1Read every tool description the model will see
- STEP 2Look for hidden instructions or exfil requests
- STEP 3Test how the agent behaves with the server connected
- STEP 4Fix: review, pin and hash tool definitions
⟩ arsenal
Tools & frameworks
MCP Inspector
MCP Server Review
mcp-scan
MCP Risk Scanner
Burp Suite Pro
Web Security
Promptfoo
Prompt Evaluation
Semgrep
Static Analysis
⟩ engage
Need mcp server pentesting expertise?
⟩ related
More in AI/ML Security
AI & GenAI Security
6+ years pioneering AI/GenAI security assessments, uncovering novel vulnerabilities in LLMs and securing AI-driven architectures.
OWASP LLM Top 10 Testing
Structured testing of LLM applications against every risk in the OWASP Top 10 for LLM Applications (2025).
LLM Red Teaming
Systematic prompt injection, jailbreak and data exfil testing.
Model Supply Chain
Securing model artifacts, weights and inference pipelines.
RAG Pipeline Security
Hardening retrieval-augmented architectures end-to-end.