Lexi - AI Legal Assistant

[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_153":154,"_155":156},"routes/_app.x.$handle.$name",{"_95":94,"_32":45,"_103":85,"_27":104,"_13":14,"_105":-5,"_11":106,"_107":108,"_109":110,"_111":112,"_113":114,"_20":-5,"_115":-5,"_116":-5,"_117":25,"_118":119,"_120":57,"_121":-5,"_122":-5,"_123":124},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":-5,"_18":-5},"project",{"_43":44,"_32":45,"_38":46,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":-5,"_53":-5,"_18":54,"_55":31,"_56":57,"_58":57,"_59":-5,"_60":61,"_62":63,"_64":-5,"_65":-5,"_66":-5,"_67":-5,"_68":69,"_70":69,"_71":72,"_73":72,"_74":72,"_75":69,"_76":69,"_77":78,"_79":80,"_81":82,"_83":25},"version",11,"changes",[19],"aiChatInfo","userId",{"_20":21,"_22":23,"_24":25,"_26":25,"_27":28,"_29":28},"file",{"_30":31,"_32":28,"_33":34,"_35":37,"_38":39,"_40":14,"_41":42,"_24":25},"previous",{"_30":31,"_32":28,"_33":34,"_35":36},"deleted",false,"deletedHere","path","main.tsx","latestPath","fileId","919501be-4154-11f0-8a14-76b3cceeab13","name","fileType","http","content","const legalTaskSuggestionSystemPromptTemplate = `\nYou are an AI assistant specializing in legal document review.\nYour goal is to suggest a comprehensive and diverse list of potential legal tasks or analyses\nthat a user might want to perform on the provided document text.\nFocus on common legal review objectives like information extraction (e.g., parties, dates, definitions, obligations, rights), summarization, compliance checks (e.g., specific regulations if inferable), risk identification, clause analysis (e.g., termination, liability, indemnification, confidentiality), and comparisons if multiple documents/sections were hypothetically present.\nThe list should be extensive to maximize the chance of including the user's intended task. Aim for at least 10-15 diverse suggestions if the document is rich enough, but fewer if the document is very short or simple.\n\nReturn your suggestions STRICTLY as a JSON array of strings ONLY. Each string should be a concise description of a task.\nExample: [\"Identify all parties mentioned in the document\", \"Extract all defined terms and their definitions\", \"Summarize the key obligations for Party A\", \"List all clauses related to termination\", \"Identify renewal conditions and notice periods\", \"Assess potential risks related to data privacy clauses\", \"Find all references to governing law and jurisdiction\", \"Extract payment terms and schedules\", \"Check for assignment clauses and their conditions\", \"Identify any force majeure provisions\"]\n\nDOCUMENT TEXT:\n==============\n%%DOCUMENT_TEXT%%\n`;\nconst legalAIAnalysisSystemPromptTemplate = `\n<Prompt Template>\nYou are an AI System for Precise Legal Information Extraction.\nYour primary goal is to perform the user-specified legal information extraction or analysis task, described below,\non the provided legal text (which will be the main content of the user message).\nYou must structure ALL your outputs strictly according to the defined shared JSON schema.\n\nLEGAL TASK / QUERY TO ADDRESS: %%LEGAL_TASK_QUERY%%\nYou will analyze the document text provided in the user message.\nThe document is described as: %%INPUT_SOURCE_DESCRIPTION%%\n\nYour expertise lies in understanding legal terminology, document structures, and nuanced phrasing.\nMeticulously analyze the input in the context of the LEGAL TASK / QUERY and then funnel your findings\ninto the generic JSON schema detailed below.\n\nSteps You Must Follow:\nUnderstand the Request: Fully grasp the LEGAL TASK / QUERY. Determine the specific information to be extracted or the analysis to be performed.\nProcess the Input: Carefully review and parse the entire document text provided in the user message. Identify all relevant sections, clauses, sentences, and data points pertinent to the LEGAL TASK / QUERY.\nPerform Legal Analysis (Reasoning): Execute the core extraction or analytical task. This involves:\nIdentifying specific pieces of information (e.g., names, dates, definitions, obligations, rights, conditions).\nInterpreting the meaning of relevant legal clauses or statements.\nSynthesizing information if the query requires it (e.g., summarizing, comparing).\nAnswering specific questions based only on the provided document text.\nStructure the Output (Conclusion): Map your analysis, findings, and conclusions precisely into the mandatory JSON output format provided below. Ensure every field in the JSON schema is considered and populated appropriately.\nMaintain Neutrality and Accuracy: Present findings objectively based on the text. If the input is insufficient or ambiguous for a particular part of the query, note this in the limitations_or_warnings field.\nOutput Format:\nThe output MUST be a single JSON object, without any markdown formatting (e.g., no \\`\\`\\`json ... \\`\\`\\` wrapper). The JSON schema is:\n\nALL FINDINGS MUST BE RETURNED. do NOT trim the list, return all findings.\n\nJSON\n\n{\n \"request_details\": {\n \"operation_performed\": \"string\",\n \"input_source_description\": \"string\"\n },\n \"analysis_summary\": {\n \"overview\": \"string\",\n \"key_findings_count\": \"integer\"\n },\n \"detailed_findings\": [\n {\n \"finding_id\": \"string\",\n \"finding_type\": \"string\",\n \"description\": \"string\",\n \"severity_or_importance\": \"string\",\n \"relevant_input_excerpt\": \"string\",\n \"page_or_section_reference\": \"string\",\n \"implications\": \"string\",\n \"confidence_score\": \"float\"\n }\n ],\n \"execution_metadata\": {\n \"status\": \"string\",\n \"timestamp\": \"string\",\n \"limitations_or_warnings\": \"string\",\n \"disclaimer\": \"string\"\n }\n}\nJSON Schema Field Explanations:\n\nrequest_details:\noperation_performed: (string) Re-state or summarize the LEGAL TASK / QUERY you performed.\ninput_source_description: (string) The description of the input document as provided in the prompt (%%INPUT_SOURCE_DESCRIPTION%%).\nanalysis_summary:\noverview: (string) A concise, high-level summary of your overall analysis pertaining to the query.\nkey_findings_count: (integer) The total number of discrete findings in detailed_findings.\ndetailed_findings: (Array of objects) Each object is a distinct finding relevant to the query.\nfinding_id: (string) A unique ID for this finding (e.g., \"F001\", \"TERM002\", \"OBLG003\").\nfinding_type: (string) Nature of the finding (e.g., \"ClauseAnalysis\", \"Obligation\", \"Right\", \"DefinedTerm\", \"RiskAssessment\", \"QuerySpecificAnswer\", \"PartyIdentification\", \"DateExtraction\").\ndescription: (string) Detailed explanation of the finding.\nseverity_or_importance: (string, optional) e.g., \"High\", \"Medium\", \"Low\", \"KeyInformation\", \"Critical\".\nrelevant_input_excerpt: (string, optional) A direct quote from the input text that supports this finding (max 500 chars). Keep it concise and directly relevant.\npage_or_section_reference: (string, optional) Specific location in the input if identifiable (e.g., \"Section 3.1a\", \"Page 5, Paragraph 2\").\nimplications: (string, optional) Potential consequences or implications of this finding, based on the text.\nconfidence_score: (float, optional, 0.0 to 1.0) Your confidence in the accuracy and relevance of this finding.\nexecution_metadata:\nstatus: (string) Should be \"Completed\", \"Completed with Limitations\", or \"Error\" (if you couldn't process the request fundamentally).\ntimestamp: (string) Generate the current ISO 8601 timestamp (e.g., \"2025-06-04T10:30:00Z\") upon completion.\nlimitations_or_warnings: (string) Any limitations faced during analysis (e.g., \"Input text was ambiguous in section X,\" \"Analysis limited to provided text only,\" \"Query partially addressed due to missing information in text\") or other warnings.\ndisclaimer: (string) Standard disclaimer: \"This AI-generated analysis is for informational purposes only and does not constitute legal advice. Consult with a qualified legal professional for advice on specific legal matters.\"\nEnsure all required string fields are populated. For optional fields, if no information is found or applicable, they can be omitted or explicitly set to null (for non-string types where appropriate). The detailed_findings array can be empty if no relevant findings are made for the specific query.\n`;\ninterface LogEntry {\n agent: string;\n message: string | object;\n type: \"step\" | \"result\" | \"error\" | \"input\" | \"final\" | \"info\";\n}\ninterface LegalAIResponse {\n request_details: { operation_performed: string; input_source_description: string };\n analysis_summary: { overview: string; key_findings_count: number };\n detailed_findings: Array<\n {\n finding_id: string;\n finding_type: string;\n description: string;\n severity_or_importance?: string;\n relevant_input_excerpt?: string;\n page_or_section_reference?: string;\n implications?: string;\n confidence_score?: number;\n }\n >;\n execution_metadata: { status: string; timestamp: string; limitations_or_warnings: string; disclaimer: string };\n}\ninterface LexiDocument {\n id: string;\n name: string;\n text: string;\n sourceType: \"url\" | \"text\" | \"file\";\n sourceValue: string;\n dateAdded: string;\n}\ninterface LexiAnalysis {\n id: string;\n docId: string;\n query: string;\n result: LegalAIResponse;\n datePerformed: string;\n}\nfunction generateHtmlShell(initialUrl, initialText, initialQuery, sourceUrl) {\n const escapedUrl = initialUrl.replace(/\"/g, \""\");\n const escapedText = initialText.replace(//g, \">\");\n const escapedQuery = initialQuery.replace(//g, \">\");\n return `\n\n\n\n \n \n Lexi - AI Legal Assistant\n \n\n\n

\n \n \n \n

\n\n\n`;\n}\nexport default async function(req: Request) {\n const { OpenAI } = await import(\"https://esm.town/v/std/openai\");\n const { fetch } = await import(\"https://esm.town/v/std/fetch\");\n const { PDFExtract, PDFExtractOptions } = await import(\"npm:pdf.js-extract\");\n const CORS_HEADERS = {\n \"Access-Control-Allow-Origin\": \"*\",\n \"Access-Control-Allow-Methods\": \"POST, GET, OPTIONS\",\n \"Access-Control-Allow-Headers\": \"Content-Type, Accept\",\n };\n if (req.method === \"OPTIONS\") { return new Response(null, { status: 204, headers: CORS_HEADERS }); }\n const url = new URL(req.url);\n const format = url.searchParams.get(\"format\");\n const action = url.searchParams.get(\"action\");\n const sourceUrl = import.meta.url.replace(\"esm.town\", \"val.town\");\n const openai = new OpenAI();\n const MAX_TEXT_LENGTH_SUGGESTION = 20000;\n const MAX_TEXT_LENGTH_ANALYSIS = 30000;\n async function extractPdfTextNative(data: ArrayBuffer, fileName: string, log: LogEntry[]): Promise {\n const agent = \"PDF Extraction Agent (Val Server)\";\n log.push({ agent, type: \"step\", message: `Processing PDF: ${fileName} using npm:pdf.js-extract` });\n try {\n const pdfExtract = new PDFExtract();\n const options: PDFExtractOptions = {};\n const pdfData = await pdfExtract.extractBuffer(data, options);\n const text = pdfData.pages.map(page => page.content.map(item => item.str).join(\" \")).join(\"\\\\n\\\\n\");\n if (!text || text.trim().length === 0) {\n log.push({ agent, type: \"error\", message: \"PDF extraction resulted in empty text content.\" });\n return null;\n }\n log.push({ agent, type: \"info\", message: `Successfully extracted ~${text.length} characters from PDF.` });\n return text;\n } catch (error) {\n const errorMessage = `Failed to extract text from PDF: ${error.message}`;\n log.push({ agent, type: \"error\", message: errorMessage });\n console.error(agent, errorMessage, error);\n return null;\n }\n }\n async function callOpenAI(\n openaiInstance: OpenAI,\n systemPrompt: string,\n userMessage: string,\n model = \"gpt-4o\",\n expectJson = true,\n log: LogEntry[],\n agentName: string,\n ): Promise

Diff · v11