Tool Parameter Injection

Inject malicious parameters into tool calls to escalate privileges or bypass validation

Tactic

Privilege Escalation · Stage 4

Escalate capabilities beyond declared scope or bypass authorization

Attack class

MCP-EXPLOIT

Attacking Model Context Protocol server configurations, tool registrations, and inter-server trust

Evidence grade

Validated

Reproduced in a controlled lab environment (DVAA) with documented steps.

DVAA validation

MCP parameter injection

Reproductions in Damn Vulnerable AI Agent, the OpenA2A intentionally-broken agent for kill-chain validation.

Honeypot

AgentPwn coverage

Live

mcp-exploitationagentpwn.com/learn ↗

An AgentPwn trap page produces a payload tagged with this technique class. Following the AgentPwn taxonomy of trap pages shows what an agent encounters.

Tool-injection tiers smuggle malicious parameters into tool calls.

Detect

Detection · HackMyAgent

Live1 live · 0 queued

PERM-001

npx hackmyagent secure --ciLive = implemented in hackmyagent; queued = declared

Defend

Defense · OASB controls

Live5 live · 0 queued

OASB 2.1 OASB 2.2 OASB 2.3 OASB 2.4 OASB 2.5

Live = documented at oasb.ai; queued = declared

Reference

How to cite

AI Agent Threat Matrix T-4003 (Tool Parameter Injection). OpenA2A, 2026. https://threats.opena2a.org/techniques/T-4003

← Back to the matrix