export interface ResearchResult {
learnings: string[];
visitedUrls: string[];
}
// ===========================
// JSON Schema (plain objects) for Gemini responses
// ===========================
export const SerpQueriesJsonSchema = {
type: 'array',
items: {
type: 'object',
properties: {
query: { type: 'string' },
researchGoal: { type: 'string' }
},
required: ['query'],
additionalProperties: false,
},
minItems: 1
} as const;
export const OutlineJsonSchema = {
type: 'object',
properties: {
outline: { type: 'array', items: { type: 'string' } }
},
required: ['outline'],
additionalProperties: false,
} as const;
export const SectionsJsonSchema = {
type: 'object',
properties: {
sections: { type: 'array', items: { type: 'string' } },
citations: { type: 'array', items: { type: 'string' } }
},
required: ['sections'],
additionalProperties: false,
} as const;
export const SummaryJsonSchema = {
type: 'object',
properties: { summary: { type: 'string' } },
required: ['summary'],
additionalProperties: false,
} as const;
export const TitleJsonSchema = {
type: 'object',
properties: { title: { type: 'string' } },
required: ['title'],
additionalProperties: false,
} as const;
export const FeedbackResponseJsonSchema = {
type: 'object',
properties: {
followUpQuestions: { type: 'array', items: { type: 'string' } },
analysis: { type: 'string' },
confidenceScore: { type: 'number' }
},
required: ['followUpQuestions', 'analysis'],
additionalProperties: false,
} as const;
export interface MCPResearchResult {
content: { type: "text"; text: string }[];
metadata: {
learnings: string[];
visitedUrls: string[];
stats: {
totalLearnings: number;
totalSources: number;
};
};
}
// ==========================================
// Centralized types and schemas (no runtime bloat with type-only imports)
// ==========================================
// Zod is used for JSON schema enforcement across analyzer/synthesizer/supervisor
import { z } from 'zod';
// Core primitives
export type EmbeddingVector = number[];
export interface Chunk {
id: string;
text: string;
tokens: number;
embedding?: EmbeddingVector;
source?: string;
}
export interface SerpQuery {
query: string;
researchGoal?: string;
}
export type Citation = string;
// Outline / Sections / Summary
export const OutlineSchema = z.object({
title: z.string(),
sections: z.array(z.string()),
}).strict();
export type Outline = z.infer<typeof OutlineSchema>;
export const SectionsSchema = z.object({
sections: z.array(z.string()),
citations: z.array(z.string()).optional(),
}).strict();
export type SectionsJSON = z.infer<typeof SectionsSchema>;
export const SummarySchema = z.object({
summary: z.string(),
}).strict();
export type SummaryJSON = z.infer<typeof SummarySchema>;
export interface Report {
abstract: string;
toc: string[];
introduction: string;
body: string[];
methodology: string;
limitations: string;
keyLearnings: string[];
references: string[];
}
// Agent IO contracts
export interface GatherOutput {
queries: string[];
chunks: Chunk[];
sources: string[];
}
export interface AnalyzeOutput {
findings: string[];
citations: Citation[];
}
export interface SynthesizeOutput {
outline: Outline;
sections: string[];
citations?: Citation[];
}
export interface SuperviseVerdict {
ok: boolean;
reasons?: string[];
repaired?: SynthesizeOutput;
}
export interface FullResearchResult {
report: Report;
meta: {
timings?: Record<string, number>;
tokens?: Record<string, number>;
};
}
// Providers interfaces
export interface EmbeddingsProvider {
embed(text: string): Promise<EmbeddingVector>;
countTokens(text: string): Promise<number>;
}
export interface LlmJsonClient<TConfig = unknown> {
generate<T>(args: { prompt: string; schema: z.ZodType<T>; config?: TConfig }): Promise<T>;
}
