MCP Connector Capability Attestation Absent

← Agentic AI (OWASP Agentic Top 10)

mcp_capability_attestation_absent   HARD

Description

A connected MCP server self-declares the capabilities (scopes) it holds, and the agent host grants them from the declaration alone with no attestation, so a connector that lists a privileged scope it was never issued is trusted with it (MCP capability-attestation gap, arXiv 2601.17549).

How it works

An MCP connector advertises the capabilities (scopes) it holds as part of its handshake, and the agent host authorizes tool calls from that self-declared list alone. Because the host performs no attestation, it cannot tell a capability it actually issued from one the connector simply claims, so a connector that lists a privileged scope it was never granted is authorized to use it and the privileged tool call executes. This is a protocol-level trust gap distinct from tool-description poisoning: the untrusted input is the capability claim itself, not a hidden instruction. The hardened host issues a signed capability grant when it provisions a connector and, at call time, verifies each requested scope against that attested grant, refusing any scope it did not issue.

How to exercise it. DVMA is the harness - open this module from the home index and tap the demo action. The screen ships the malicious input and simulates the attacker (e.g. the companion app, crafted intent, or scanned payload) in-process, and the evidence panel prints the proof. The Tools (optional) and Attack inputs below are only needed to reproduce the exploit end-to-end on a real device.

Exploit steps

1. Set up. Build DVMA with a flavor that enables the Agentic AI (OWASP Agentic Top 10) category (e.g. --dart-define-from-file=config/flavors/dev.json) and run on an emulator/simulator you control. The demo needs no external tooling; for the optional on-device reproduction the relevant tools are: promptfoo, MCP scanner, Burp Suite, mitmproxy.
2. Locate the target. From the home index, open MCP Connector Capability Attestation Absent (mcp_capability_attestation_absent). The How it works section above describes this module’s specific weakness; the screen states the intended-secure behavior and exposes the vulnerable action.
3. Exploit. Interact with the agent’s memory / tool / sub-agent channel, plant the malicious content, and confirm it influences a later action or is acted on without authentication.
4. Observe the evidence. Trigger the vulnerable action and read the evidence panel - it prints the concrete proof (leaked value, accepted replay, executed payload, or unauthorized result).
5. Contrast with the secure path. Run the module’s secure/hardened action (where provided) and confirm the same attack is rejected - this is what a correct implementation should do.

Tools (optional)

• promptfoo
• MCP scanner
• Burp Suite
• mitmproxy

garak / promptfoo / MCP scanner test the LLM or MCP endpoint behind the app, not the app binary. Point them at the backend model API (find it with mitmproxy / Burp Suite) or a local on-device model server; the in-app demo already exercises the same prompt path.

Attack inputs

Payloads/artifacts you author for the on-device attack. The demo already ships and simulates these in-process (e.g. the malicious companion app / crafted intent is emulated inside the screen), so you only need to craft them to reproduce the exploit on a real device:

• connector declaring unattested capabilities

Real-world references

Concrete public disclosures that match this vulnerability class:

• Breaking the Protocol: absence of MCP capability attestation
• CWE-290 (Authentication Bypass by Spoofing)