JuliusBrussee/caveman · error · Error

caveman agent: Standard Schema needs inputJSONSchema or Stan

Error message

caveman agent: Standard Schema needs inputJSONSchema or Standard JSON Schema conversion

What it means

For a Standard Schema tool input, tool() needs a JSON Schema to send to the model. It first checks options.inputJSONSchema, then tries the standard.jsonSchema converter. If neither route yields a schema (converted is still undefined — the schema object exposes no jsonSchema converter and none was supplied), it throws 'Standard Schema needs inputJSONSchema or Standard JSON Schema conversion'.

Source

Thrown at packages/agent/src/primitives.ts:162

  const standard = standardToolSchema(options.input);
  let input: TSchema;
  if (standard === undefined) {
    input = options.input as TSchema;
  } else {
    let converted = "inputJSONSchema" in options
      ? options.inputJSONSchema
      : undefined;
    if (converted === undefined && standard.jsonSchema !== undefined) {
      try {
        converted = standard.jsonSchema.input({ target: "draft-07" });
      } catch (error) {
        throw new Error("caveman agent: Standard Schema cannot emit draft-07 input JSON Schema", {
          cause: error,
        });
      }
    }
    if (converted === undefined) {
      throw new Error(
        "caveman agent: Standard Schema needs inputJSONSchema or Standard JSON Schema conversion",
      );
    }
    if (!isRecord(converted)) {
      throw new Error("caveman agent: Standard Schema emitted invalid input JSON Schema");
    }
    input = converted as TSchema;
  }
  const definition = {
    kind: "tool",
    name: options.name,
    description: options.description,
    input,
    effect: options.effect,
    result,
    ...(typeof options.result === "object" ? { artifact: options.result } : {}),
    ...(options.allowRepeat === undefined ? {} : { allowRepeat: options.allowRepeat }),
    timeoutMs,

View on GitHub (pinned to 27d5a3981a)

Solutions

1. Add inputJSONSchema alongside input with the draft-07 schema for the model
2. Upgrade the schema library to a version that ships a Standard JSON Schema converter on ~standard
3. Or switch input to a TypeBox/native TSchema, which this library consumes directly

Example fix

// before
const t = tool({ name: "echo", effect: "read", input: customStandardSchema, execute: ... }); // no converter

// after
const t = tool({
  name: "echo",
  effect: "read",
  input: customStandardSchema,
  inputJSONSchema: { type: "object", properties: { msg: { type: "string" } } },
  execute: ...,
});

const std = input["~standard"];
if (!std || std.jsonSchema?.input === undefined) {
  if (!inputJSONSchema) {
    throw new Error("schema has no JSON Schema converter; provide inputJSONSchema");
  }
}

const hasJsonSchemaConverter = (
  v: unknown,
): v is { "~standard": { jsonSchema: { input: unknown } } } => {
  const std = (v as { "~standard"?: { jsonSchema?: { input?: unknown } } })["~standard"];
  return std?.jsonSchema?.input !== undefined;
};

When it happens

Trigger: Passing a minimal Standard Schema implementation that has ~standard.validate but no jsonSchema converter property, and not providing inputJSONSchema. Note this is the no-converter case; a converter that throws produces error 175 instead.

Common situations: Hand-rolled schema wrappers, lightweight validators that implement only the validation half of Standard Schema, or an older schema library version released before the JSON Schema converter extension existed.

Related errors

• caveman agent: Standard Schema cannot emit draft-07 input JS
• caveman agent: Standard Schema emitted invalid input JSON Sc
• cave_execution_authorization_private
• caveman agent: file path is required
• caveman agent: invalid tool name ${JSON.stringify(options.na

AI-assisted analysis of JuliusBrussee/caveman@27d5a3981a (2026-08-15). Data as JSON: /api/errors/91018b88eb0cb539. Report an issue: GitHub.