OrbiResearch
A security research laboratory focused on credential management and identity security for AI agents, which proposed a short-lived plus bound-runtime security model for agent credentials.
Timeline (last 90 days)
The OrbiResearch lab named the attack method caused by credentials being replayed, cloned, or used beyond expiry the "identity forgery pattern."
OrbiResearch proposes that Agent credentials should be short-lived by default, issued per session rather than per Agent, generated at run startup and dying at run end.
For long-running Agents that legitimately run for hours or even days, OrbiResearch's approach is not to extend the expiry window, but to rotate credentials at fixed intervals within the run, with the rule that no credential's overdue survival time should exceed several times the normal task duration.
All Facts (3)
The OrbiResearch lab named the attack method caused by credentials being replayed, cloned, or used beyond expiry the "identity forgery pattern."
50%UnverifiedOrbiResearch proposes that Agent credentials should be short-lived by default, issued per session rather than per Agent, generated at run startup and dying at run end.
50%UnverifiedFor long-running Agents that legitimately run for hours or even days, OrbiResearch's approach is not to extend the expiry window, but to rotate credentials at fixed intervals within the run, with the rule that no credential's overdue survival time should exceed several times the normal task duration.
50%