The fastest, cheapest, most efficient open-source agent harness. Run more than 500 agents on a $10 VPS.
Rust agent-orchestrationai-agentsai-assistantdesktopllmlocal-firstmcppersonal-aiprivacyrustsecond-braintauri
fix(tinyagents): capture failure cause before agent finishes
Steven Enamakela0836a2b23fd
diff --git a/crates/openhuman-core/src/agent/tinyagents/middleware/repeated_failure.rs b/crates/openhuman-core/src/agent/tinyagents/middleware/repeated_failure.rsindex cc77e72585..4b9f901422 100644--- a/crates/openhuman-core/src/agent/tinyagents/middleware/repeated_failure.rs+++ b/crates/openhuman-core/src/agent/tinyagents/middleware/repeated_failure.rs@@ -532,6 +532,11 @@ impl Middleware<(), OpenHumanRunContext> for RepeatedToolFailureMiddleware {state: &(),run: &mut tinyagents_harness::middleware::AgentRun,) -> TaResult<()> {+ // The owner's `after_agent` finishes the run and clears the stop latch,+ // so read the typed cause first or it is lost on a classified halt.+ if run.failure_cause.is_none() {+ run.failure_cause = self.policy.stop.current_classified_cause();+ }self.owner.after_agent(ctx, state, run).await}