ai-dispatch 9.3.2

Multi-AI CLI team orchestrator
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110

// Detects "model unavailable" agent failures (deprecated/renamed/unsupported model ids).
// Exports: is_model_unavailable_error, extract_model_unavailable_message.
// Deps: serde_json for parsing agent JSON error events.

/// True when `message` indicates the requested model id is not usable —
/// deprecated, renamed, unknown, or unsupported for this account. Distinct from
/// rate-limit/quota (transient) errors: this is fixed by choosing another model.
pub fn is_model_unavailable_error(message: &str) -> bool {
    let lower = message.to_lowercase();
    lower.contains("model not found")
        || lower.contains("not supported model")
        || lower.contains("model_not_found")
        || lower.contains("unknown model")
        || lower.contains("invalid model")
        || lower.contains("no such model")
        || lower.contains("unsupported model")
        || (lower.contains("model") && lower.contains("is not supported"))
        || (lower.contains("model") && lower.contains("does not exist"))
        || (lower.contains("model") && lower.contains("is not available"))
        || (lower.contains("model") && lower.contains("has been deprecated"))
        || (lower.contains("model") && lower.contains("no longer available"))
}

/// Pull a model-unavailable message out of one raw output line. Handles both
/// plain text and the JSON error events that PTY agents (opencode/mimo/codex)
/// stream, mirroring `rate_limit::extract_rate_limit_message`.
pub fn extract_model_unavailable_message(raw: &str) -> Option<String> {
    let trimmed = raw.trim();
    if trimmed.is_empty() {
        return None;
    }
    if trimmed.starts_with('{') && trimmed.contains("\"error\"") {
        if let Some(msg) = extract_from_json_error(trimmed) {
            return Some(msg);
        }
    }
    if is_model_unavailable_error(trimmed) && trimmed.len() < 500 {
        Some(trimmed.to_string())
    } else {
        None
    }
}

fn extract_from_json_error(json_str: &str) -> Option<String> {
    let value: serde_json::Value = serde_json::from_str(json_str).ok()?;
    // Walk common nesting shapes: {message}, {error:{message}}, {error:{data:{message}}}.
    for candidate in [
        value.get("message"),
        value.get("error").and_then(|e| e.get("message")),
        value.get("error").and_then(|e| e.get("data")).and_then(|d| d.get("message")),
    ] {
        if let Some(msg) = candidate.and_then(serde_json::Value::as_str) {
            let msg = msg.trim();
            if !msg.is_empty() && is_model_unavailable_error(msg) {
                return Some(msg.to_string());
            }
        }
    }
    None
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn detects_real_agent_failures() {
        // opencode
        assert!(is_model_unavailable_error("Model not found: glm-4.7/."));
        // mimo
        assert!(is_model_unavailable_error("Not supported model mimo-v2.5-pro-ultraspeed: Param Incorrect"));
        // codex
        assert!(is_model_unavailable_error(
            "The 'gpt-4.1-nano' model is not supported when using Codex with a ChatGPT account."
        ));
    }

    #[test]
    fn ignores_unrelated_and_transient_errors() {
        assert!(!is_model_unavailable_error("rate limit exceeded"));
        assert!(!is_model_unavailable_error("Insufficient balance. Manage your billing"));
        assert!(!is_model_unavailable_error("Authentication failed. Please log in"));
        assert!(!is_model_unavailable_error("connection reset by peer"));
    }

    #[test]
    fn extracts_from_json_error_events() {
        let opencode = r#"{"type":"error","error":{"name":"UnknownError","data":{"message":"Model not found: glm-4.7/."}}}"#;
        assert_eq!(
            extract_model_unavailable_message(opencode).as_deref(),
            Some("Model not found: glm-4.7/.")
        );
        let codex = r#"{"type":"error","status":400,"error":{"type":"invalid_request_error","message":"The 'gpt-4.1-nano' model is not supported when using Codex with a ChatGPT account."}}"#;
        assert!(extract_model_unavailable_message(codex).is_some());
    }

    #[test]
    fn ignores_json_rate_limit_events() {
        let rl = r#"{"type":"error","error":{"message":"429 rate limit exceeded"}}"#;
        assert!(extract_model_unavailable_message(rl).is_none());
    }

    #[test]
    fn extracts_from_plain_text_line() {
        assert_eq!(
            extract_model_unavailable_message("Error: unknown model 'foo'").as_deref(),
            Some("Error: unknown model 'foo'")
        );
    }
}