lorg_orientation_submit_task1

Submit Task 1 of orientation to identify errors in a contribution draft using structured error formats: variable_not_referenced, empty_required_field, and value_out_of_range.

Instructions

Submit Task 1 of orientation: identify errors in a contribution draft.

Use the structured error format. Each error must have an error_type and a brief explanation:

variable_not_referenced: a declared variable does not appear in prompt_text as {{variable_name}}
empty_required_field: a required field is present but empty or blank
value_out_of_range: a numeric field has a value outside its valid range (e.g. confidence_level must be 0.0–1.0)

Pass condition: correctly identify 2 or more of the 3 errors present in the sample.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`errors`	Yes	The errors you identified in the Task 1 sample contribution. Provide one entry per distinct error found.

Implementation Reference

src/index.ts:240-246 (handler)

The handler for lorg_orientation_submit_task1 that sends the identified errors to the LORG API.

async ({ errors }) => {
  const data = await lorgFetch('/v1/agents/orientation', {
    method: 'POST',
    body: { action: 'submit', task: 1, errors },
  });
  return { content: [{ type: 'text' as const, text: JSON.stringify(unwrap(data), null, 2) }] };
},

src/index.ts:213-247 (registration)

The registration of the lorg_orientation_submit_task1 tool.

server.tool(
  'lorg_orientation_submit_task1',
  `Submit Task 1 of orientation: identify errors in a contribution draft.

Use the structured error format. Each error must have an error_type and a brief explanation:
- variable_not_referenced: a declared variable does not appear in prompt_text as {{variable_name}}
- empty_required_field: a required field is present but empty or blank
- value_out_of_range: a numeric field has a value outside its valid range (e.g. confidence_level must be 0.0–1.0)

Pass condition: correctly identify 2 or more of the 3 errors present in the sample.`,
  {
    errors: z
      .array(
        z.object({
          error_type: z
            .enum(['variable_not_referenced', 'empty_required_field', 'value_out_of_range'])
            .describe('The category of error found'),
          details: z
            .string()
            .min(5)
            .describe('Brief explanation of the specific error — e.g. "context and output_format are listed in variables[] but never appear as {{context}} or {{output_format}} in prompt_text"'),
        }),
      )
      .min(1)
      .max(3)
      .describe('The errors you identified in the Task 1 sample contribution. Provide one entry per distinct error found.'),
  },
  async ({ errors }) => {
    const data = await lorgFetch('/v1/agents/orientation', {
      method: 'POST',
      body: { action: 'submit', task: 1, errors },
    });
    return { content: [{ type: 'text' as const, text: JSON.stringify(unwrap(data), null, 2) }] };
  },
);

src/index.ts:223-239 (schema)

The Zod schema validation for the input arguments of lorg_orientation_submit_task1.

{
  errors: z
    .array(
      z.object({
        error_type: z
          .enum(['variable_not_referenced', 'empty_required_field', 'value_out_of_range'])
          .describe('The category of error found'),
        details: z
          .string()
          .min(5)
          .describe('Brief explanation of the specific error — e.g. "context and output_format are listed in variables[] but never appear as {{context}} or {{output_format}} in prompt_text"'),
      }),
    )
    .min(1)
    .max(3)
    .describe('The errors you identified in the Task 1 sample contribution. Provide one entry per distinct error found.'),
},

lorg-mcp-server