tinyhumansai/openhuman · error

Prompt blocked by security policy.

Error message

Prompt blocked by security policy.

What it means

Error "Prompt blocked by security policy." thrown in tinyhumansai/openhuman.

Source

Thrown at src/openhuman/agent/harness/session/runtime.rs:803

    /// Runs a single turn with the given message and returns the response.
    ///
    /// This is the primary high-level method for programmatic interaction with the agent.
    /// It wraps the core `turn` logic with telemetry events (`AgentTurnStarted`,
    /// `AgentTurnCompleted`) and error sanitization.
    pub async fn run_single(&mut self, message: &str) -> Result<String> {
        let guard = enforce_prompt_input(
            message,
            PromptEnforcementContext {
                source: "agent.runtime.run_single",
                request_id: None,
                user_id: Some(self.event_channel()),
                session_id: Some(self.event_session_id()),
            },
        );
        if !matches!(guard.action, PromptEnforcementAction::Allow) {
            let user_message = match guard.action {
                PromptEnforcementAction::Allow => "Message accepted.",
                PromptEnforcementAction::Blocked => "Prompt blocked by security policy.",
                PromptEnforcementAction::ReviewBlocked => {
                    "Prompt flagged for security review and was not processed."
                }
            };
            let action_tag = match guard.action {
                PromptEnforcementAction::Allow => "allow",
                PromptEnforcementAction::Blocked => "blocked",
                PromptEnforcementAction::ReviewBlocked => "review_blocked",
            };
            crate::core::observability::report_error(
                user_message,
                "agent",
                "prompt_injection_blocked",
                &[
                    ("session_id", self.event_session_id()),
                    ("channel", self.event_channel()),
                    ("action", action_tag),
                ],

View on GitHub (pinned to a221052e0d)

When it happens

Trigger: Thrown at src/openhuman/agent/harness/session/runtime.rs:803 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tinyhumansai/openhuman@a221052e0d (2026-08-16). Data as JSON: /api/errors/a91257781b1dce85. Report an issue: GitHub.