aaif-goose/goose · error

instructions' is not a string

Error message

instructions' is not a string

What it means

Thrown during recipe response parsing when the JSON object contains "instructions" but its value is not a string (as_str() fails). The model returned instructions as an object, array, or number instead of a plain string.

Source

Thrown at crates/goose/src/agents/agent.rs:3847

            content.len()
        );

        // the response may be contained in ```json ```, strip that before parsing json
        let re = Regex::new(r"(?s)```[^\n]*\n(.*?)\n```").unwrap();
        let clean_content = re
            .captures(&content)
            .and_then(|caps| caps.get(1).map(|m| m.as_str()))
            .unwrap_or(&content)
            .trim()
            .to_string();

        let (instructions, activities) =
            if let Ok(json_content) = serde_json::from_str::<Value>(&clean_content) {
                let instructions = json_content
                    .get("instructions")
                    .ok_or_else(|| anyhow!("Missing 'instructions' in json response"))?
                    .as_str()
                    .ok_or_else(|| anyhow!("instructions' is not a string"))?
                    .to_string();

                let activities = json_content
                    .get("activities")
                    .ok_or_else(|| anyhow!("Missing 'activities' in json response"))?
                    .as_array()
                    .ok_or_else(|| anyhow!("'activities' is not an array'"))?
                    .iter()
                    .map(|act| {
                        act.as_str()
                            .map(|s| s.to_string())
                            .ok_or(anyhow!("'activities' array element is not a string"))
                    })
                    .collect::<Result<_, _>>()?;

                (instructions, activities)
            } else {
                tracing::warn!("Failed to parse JSON, falling back to string parsing");

View on GitHub (pinned to 3810898a74)

Solutions

  1. Retry the generation — shape errors are often nondeterministic
  2. Switch to a model known to follow goose's recipe prompt
  3. Trim the input conversation to reduce schema drift
  4. Hand-write the recipe YAML instead of generating it
Defensive patterns

Strategy: retry

Validate before calling

// Rust — check the field type before downstream parsing
let ok = json.get("instructions").map(|i| i.is_string()).unwrap_or(false);
if !ok { /* regenerate instead of failing */ }

Try / catch

// Rust — same retry envelope as the other recipe JSON-shape errors
match generate_recipe(&agent, &conversation).await {
    Ok(r) => Ok(r),
    Err(e) if e.to_string().contains("is not a string") => retry_or_fallback(e),
    Err(e) => Err(e),
}

Prevention

When it happens

Trigger: Model returns {"instructions": {"text": "..."}} or {"instructions": ["a", "b"]} — structured rather than scalar — during recipe generation.

Common situations: Models that over-structure output; prompt templates asking for 'detailed instructions' nudging the model into nested formats; JSON-schema-tuned models exporting every field as an object.

Related errors


AI-assisted analysis of aaif-goose/goose@3810898a74 (2026-08-16). Data as JSON: /api/errors/e10638649c3a153b. Report an issue: GitHub.