Join our Newsletter — 33% off our NHI Course
Home› FAQ› Cyber Security› What is the difference between Agent Forwarding and…
Cyber Security

What is the difference between Agent Forwarding and ProxyJump for multi-hop SSH access?

← Back to all FAQ
By NHI Mgmt Group Editorial Team Updated October 31, 2026 Domain: Cyber Security

Agent Forwarding passes authentication capability through the jump host by relaying requests to the local ssh agent. ProxyJump forwards the SSH connection through an intermediary without exposing the local agent to the remote host. In practice, ProxyJump preserves the hop while reducing key exposure risk and keeping authentication control on the client side.

What the difference means in practice for multi-hop SSH

Agent Forwarding and ProxyJump both help you reach a destination through a bastion or jump host, but they solve different problems. Agent Forwarding preserves your SSH client authentication path across hops, while ProxyJump preserves the network path only. That distinction matters because the first extends trust to the intermediate host, whereas the second keeps authentication anchored on the client.

For operators, the practical question is not just whether the connection succeeds, but where the authentication capability lives during the journey. With Agent Forwarding, the remote hop can ask your local agent to sign challenges on its behalf. With ProxyJump, the intermediate host relays traffic and never needs access to your agent, which is a cleaner fit when you want to reduce exposure across untrusted hops.

The choice is often about trust boundaries. If a jump host is strictly controlled and you need convenience for chained access, Agent Forwarding can be workable. If the intermediary is shared, external, or harder to harden, ProxyJump is usually the safer default because it avoids making the remote hop part of the credential-bearing trust chain.

Why agent exposure changes the security model

Agent Forwarding expands what a compromised jump host can influence. Even though the private key is not copied, the forwarded agent can still be abused to authenticate to other systems reachable from that host, which turns a convenience feature into a lateral movement enabler if the intermediary is compromised.

ProxyJump avoids that specific exposure by keeping the local agent on the client side. The jump host sees an SSH session relay, not your signing capability, so compromise of the intermediary does not automatically create a reusable authentication path to downstream systems.

In other words, Agent Forwarding is about delegated authentication capability, while ProxyJump is about transport chaining. If the threat you care about is jump-host compromise, ProxyJump narrows the blast radius far more effectively than forwarding an agent across the hop.

When to prefer one over the other

Use Agent Forwarding only when the intermediate host is trusted enough to participate in authentication and the operational need is strong enough to justify that added exposure. For example, tightly controlled admin workflows may accept it when repeated manual key handling would create even greater operational risk.

Prefer ProxyJump when you want multi-hop access without extending authentication authority beyond the client. It is usually the better default for bastions, partner environments, and any path where the first hop is not fully trusted to handle your SSH agent safely.

For teams standardising remote access, the right design choice is often to treat forwarding as an exception and ProxyJump as the baseline. That keeps the architecture simpler to reason about, makes review easier, and reduces the chance that a convenience setting quietly broadens privilege across environments.

Risk and Threat Considerations

Agent Forwarding increases the impact of a jump-host compromise because the intermediate system can request signatures from your local agent while the session is open. That creates a lateral movement path if the bastion is abused, poorly monitored, or shared across operators.

Failure mechanism: The remote host does not obtain your private key, but it can still act as a proxy for authentication requests, which may let an attacker reuse your live SSH access to reach additional systems.

Impact: A single compromised intermediary can become a bridge to multiple internal targets, making detection and containment harder than with a relay-only design.

Standards & Framework Alignment

This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.

CIS Controls v8 and NIST SP 800-53 Rev 5 set the technical controls, while ISO/IEC 27001:2022 defines the regulatory obligations.

FrameworkControl / ReferenceRelevance
CIS Controls v8CIS-6 — Access Control ManagementSSH hop choice affects who can use delegated access.
Recommendation — Restrict forwarded SSH access to approved jump paths and trusted admins.
NIST SP 800-53 Rev 5IA-9 — Identification and Authentication (Service and Systems)Multi-hop SSH is about authenticating systems and relayed access paths.
AC-17 — Remote AccessSSH bastions are a remote-access pattern where hop design changes exposure.
Recommendation — Use system-to-system authentication controls that avoid exposing reusable client auth paths. Enforce approved remote-access pathways and limit what intermediaries can relay.
ISO/IEC 27001:2022A.5.15 — Access controlThe comparison is fundamentally about preserving access boundaries across hops.
Recommendation — Define and enforce access paths so intermediaries do not inherit unnecessary authority.

Practitioner Guidance

What to verify: Confirm whether your SSH workflow actually needs delegated authentication or only needs hop chaining. If the answer is only hop chaining, standardise on ProxyJump and reserve Agent Forwarding for narrow, reviewed exceptions.

Common mistake: Treating Agent Forwarding as harmless because the private key stays local. The practical risk is not key extraction, it is the ability of a remote host to use your active authentication context while the connection is live.

What good looks like: Bastions are used as transport relays, not as credential-intermediation points, and operator access is limited so that a compromised hop cannot silently widen access to downstream systems.

Practitioner takeaway: Choose the mechanism that preserves the minimum trust boundary needed for the task, and default to the one that keeps authentication authority on the client whenever the intermediary does not truly need it.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Reviewed and updated by the NHIMG editorial team on October 31, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org