tinyhumansai/openhuman · error · AgentError::RegistryValidationFailed
Capability registry validation failed ({} error diagnostic(s
Error message
Capability registry validation failed ({} error diagnostic(s)): {} What it means
Raised at the end of capability-registry validation when at least one diagnostic had error severity (as opposed to warnings, which are only logged non-fatally). Each error diagnostic is a malformed capability declaration — wrong kind, name, or message — collected during registry build; the turn is aborted because an invalid capability registry cannot be trusted to gate what the model is allowed to do. The message counts the diagnostics and lists them as '[kind] name: message' lines.
Source
Thrown at src/openhuman/agent/tinyagents/mod.rs:633
name = %diag.name,
"[registry] non-fatal diagnostic: {}",
diag.message
);
}
if !errors.is_empty() {
let messages: Vec<String> = errors
.iter()
.map(|d| format!("[{}] {}: {}", d.kind.as_str(), d.name, d.message))
.collect();
for msg in &messages {
tracing::error!("[registry] error-severity diagnostic aborting turn: {msg}");
}
tracing::error!(
error_count = messages.len(),
warning_count = warnings.len(),
"[registry] aborting turn before model dispatch: capability registry validation failed"
);
return Err(anyhow::Error::new(
crate::openhuman::agent::error::AgentError::RegistryValidationFailed {
diagnostics: messages,
},
));
}
tracing::debug!(
warning_count = warnings.len(),
"[registry] registry diagnostics present (warnings only); proceeding with turn"
);
}
let mut config = RunConfig::new("agent_turn")
.with_max_model_calls(max_iterations)
.with_max_tool_calls(max_iterations.saturating_mul(8).max(8))
.with_max_depth(MAX_SPAWN_DEPTH)
.with_tag("openhuman")
.with_tag(if subagent_scope.is_some() {
"scope:subagent"View on GitHub (pinned to 7491200858)
Solutions
- Read the enumerated per-diagnostic messages logged just before this error to identify which capability declarations are malformed
- Fix or remove the offending capability entries in the registry source/config that produced them
- Re-run and confirm validation completes with only non-fatal warnings
Defensive patterns
Strategy: validation
When it happens
Trigger: Thrown at src/openhuman/agent/tinyagents/mod.rs:633 when the library encounters an invalid state.
Common situations: See trigger scenarios.
AI-assisted analysis of tinyhumansai/openhuman@7491200858 (2026-08-17).
Data as JSON: /api/errors/fc33d52317f61c48.
Report an issue: GitHub.