Strumento di Compliance Check settoriale

Overview

This workflow automates the generation of a comprehensive GDPR compliance report for a specific company. It gathers data from three primary sources: web searches for public information, scraping of the company's website, and reading existing documents from a specified Google Drive folder. The collected data is analyzed for DPO requirements and DPIA necessity, merged, and transformed into a formal, legally-structured compliance document which is then saved to Google Drive.

Workflow Chain

[Input Dinamici]  -> [Web Analysis & Crawler]    -> [Fetch URL as Markdown]      -> [Web Crawler - Scansione Sito]        -> [Analisi Obbligo DPO (Art. 37 GDPR)]        -> [DPIA - Data Protection Impact Assessment]          -> [Text Merge]            -> [Documento Compliance Finale]              -> [Save File - Compliance Report]  -> [Input Nome Azienda]    -> [Web Search - Compliance]      -> [Fetch URL as Markdown]  -> [Input Sito Web]    -> [Web Search - Compliance]  -> [Input Cartella Drive (Opzionale)]    -> [Documento Report Compliance]      -> [Save File - Report Finale]

(Note: The actual data flow converges through the analysis and merging nodes to produce the final report.)

Node Reference

Input Dinamici (Section)

• What it does: Acts as a container for initial user-provided parameters.
• Inputs from: N/A
• Configurable inputs: Input Sito Web (String), Input Nome Azienda (String), Input Cartella Drive (Opzionale) (Drive Folder Reader).
• Output: Individual string and file reference values.

Web Analysis & Crawler (Section)

• What it does: Orchestrates the web-based data gathering process.
• Inputs from: Input Sito Web and Input Nome Azienda.
• Configurable inputs: N/A
• Output: Scraped web content and search results.

Web Search - Compliance (Workspace)

• What it does: Performs a Brave Search to find relevant URLs.
• Inputs from: Input Nome Azienda and Input Sito Web.
• Configurable inputs: query (String), country (Select: Any, US, GB, CA, AU, IT), num_results (Number, 1-20).
• Output: urls (Text[]), result_count (Number), rate_limited (Boolean).

Fetch URL as Markdown (Workspace)

• What it does: Scrapes the content of provided URLs and converts them to Markdown.
• Inputs from: Web Search - Compliance.
• Configurable inputs: urls (Text[]), timeout_seconds (Number, 5-120).
• Output: markdown (Text[]), title (Text[]), success_count (Number).

Web Crawler - Scansione Sito (Workspace)

• What it does: Analyzes scraped website content to identify privacy-relevant pages.
• Inputs from: Fetch URL as Markdown.
• Configurable inputs: input_text (Text), prompt (Text), llmModelId (LLM Picker).
• Output: output_text (Text) containing a summary of relevant pages.

Analisi Obbligo DPO (Art. 37 GDPR) (Workspace)

• What it does: Evaluates if the company is legally required to appoint a DPO.
• Inputs from: Web Crawler - Scansione Sito.
• Configurable inputs: input_text (Text), prompt (Text), llmModelId (LLM Picker).
• Output: output_text (Text) detailing DPO necessity.

DPIA - Data Protection Impact Assessment (Workspace)

• What it does: Assesses the need for a formal DPIA based on data processing risks.
• Inputs from: Web Crawler - Scansione Sito.
• Configurable inputs: input_text (Text), prompt (Text), llmModelId (LLM Picker).
• Output: output_text (Text) containing risk assessment and mitigation plans.

Text Merge (Workspace)

• What it does: Combines the DPO analysis and DPIA assessment into one text block.
• Inputs from: Analisi Obbligo DPO and DPIA.
• Configurable inputs: text1 (Text), text2 (Text).
• Output: text (Merged Text).

Documento Compliance Finale (Workspace)

• What it does: Transforms all gathered intelligence into a formal, structured legal document.
• Inputs from: Text Merge and Documento Report Compliance (from Drive).
• Configurable inputs: input_texts (Text[]), prompt (Text), llmModelId (LLM Picker).
• Output: output_text (The complete formal report).

Save File - Compliance Report (Workspace)

• What it does: Finalizes the process by saving the generated report to Google Drive.
• Inputs from: Documento Compliance Finale.
• Configurable inputs: text (Text), name (Text), destinationFolder (DriveItem).
• Output: driveItemId (DriveItem).

Execution Strategy

You MUST complete all 4 phases in order: Phase 1 -> Phase 2 -> Phase 3 -> Phase 4. Do NOT skip any phase.

Phase 1: Initialization & Decision

1. Call requestUserDecision to collect the following:
   • "Qual è l'URL del sito web da analizzare?" (Context: Needed for the Web Crawler).
   • "Qual è il nome esatto dell'azienda?" (Context: Needed for search queries).
   • "Seleziona una cartella Drive per il salvataggio del report finale." (Context: Required for the terminal output node).
   • "Desideri pubblicare il report direttamente dopo la generazione?" (Context: Determines if the final step is automated or requires confirmation). Phase 1 complete. Now proceed to Phase 2. Do NOT skip to Phase 3.

Phase 2: Data Gathering

2. For EACH URL provided by Web Search - Compliance, repeat: a. Execute Fetch URL as Markdown. b. If success_count > 0, proceed; else, log error. Phase 2 complete. Now proceed to Phase 3. Do NOT skip to Phase 4.

Phase 3: Analysis & Synthesis

3. Execute Web Crawler - Scansione Sito using the fetched markdown.
4. Execute Analisi Obbligo DPO (Art. 37 GDPR) using the output from step 3.
5. Execute DPIA - Data Protection Impact Assessment using the output from step 3.
6. Execute Text Merge to combine the outputs of step 4 and step 5.
7. Execute Documento Compliance Finale using the merged text and any documents retrieved from the Input Cartella Drive. Phase 3 complete. Now proceed to Phase 4. Do NOT skip to Phase 5.

Phase 4: Finalization

8. If the user chose to publish in Step 1, execute Save File - Compliance Report with the generated text.
9. If the user did NOT choose to publish, call requestUserDecision to ask: "Il report è pronto. Vuoi procedere al salvataggio su Drive?"
10. If confirmed, execute Save File - Compliance Report. Phase 4 complete.

followUp

Configuration Tips

Error Handling

Re-editable artifacts

Incremental Edits to an Existing Report

• Detection — If the user provides an existing report text instead of a new company name, treat this as an EDIT.
• Add / remove / reorder recipes — To add a new compliance section, use sectionFinder to locate the target section, sectionManager(ungroup, { sectionId }), then organizeNodes and sectionManager(group, { nodeIds, label }).
• Post-edit — After modifying the prompt or input text, re-run the Documento Compliance Finale node and honor the original publish decision.