- Home
- Skills
- Research & Analysis
- fonti primarie it
fonti primarie it
Forensic extraction and verification tools for Italian public administration documents, procurement data, and ISTAT stats.
$5
Works with the AI Agents you already use
fonti primarie it
Example session with this skill installed
Analizza il file delibera_123.pdf nella cartella fonti, estrai il testo e crea la citazione formattata per il mio dossier.
- Read your context and instructions
- Compiled the fonti primarie it
- Generated the data export
L'analisi del documento è stata completata con successo. Il sistema ha rilevato testo nativo e ha generato la citazione verificabile con il contenuto estratto per pagina.
Estrazione Atto PDF
Documento: delibera_123.pdf
Metodo: pdftotext -layout (pagine 1-2)
Data Elaborazione: 2026-11-05
| Pagina | Metodo | Anteprima Testo | Citazione Verificabile |
|---|---|---|---|
| 1 | pdftotext | COMUNE DI LODI - Deliberazione n. 123. Oggetto: Affidamento lavori... | fonti/delibera_123.pdf, p. 1 (2026-11-05) |
| 2 | pdftotext | ...visto il parere di regolarità tecnica. Il Segretario Generale... | fonti/delibera_123.pdf, p. 2 (2026-11-05) |
Testo Estratto (Pagina 1)
COMUNE DI LODI
Settore Lavori Pubblici
DELIBERAZIONE DELLA GIUNTA COMUNALE N. 123
Data: 14/03/2026
OGGETTO: Affidamento dei lavori di rifacimento del manto stradale in Via Roma.
CIG: Z1A2B3C4D5
Importo complessivo: € 248.500,00
La Giunta, presieduta dal Sindaco...
Next steps
- Usa
dati_query.pyse devi estrarre il CIG e confrontarlo con altri appalti nel CSV. - Esegui
tabelle.pyse il PDF contiene allegati tecnici con computi metrici complessi. - Archivia il file markdown generato nella cartella
dati/per il dossier finale.
delibera_123.pdf
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
This skill provides a suite of Python-based forensic tools for investigative journalists and researchers focused on Italian public administration data. It eliminates the manual drudgery of extracting verified information from government portals, PDF resolutions, and demographic databases. The toolset automates the transition from raw documents to citable evidence by integrating OCR, table extraction, and structured search.
What it does
- Document extraction reads Italian PA PDFs, automatically applying OCR to scans and citing specific page numbers.
- Table reconstruction recovers complex financial data and unified cells from PDF budgets using remote processing.
- PA scraping extracts clean text and publication dates from Italian municipal portals while archiving the original HTML as evidence.
- Demographic analysis fetches ISTAT demographic series for specific municipalities without requiring API keys.
- Semantic archiving indexes local document collections for conceptual search using local embedding models.
- Integrity verification checks historical page versions via Internet Archive and validates CSV datasets for duplicates or gaps.
How it works
- Use
atti_pdf.pyto convert local or batch PDFs into markdown with automated OCR and page-level citations. - Run
tabelle.pyon complex financial reports to reconstruct tables that standard parsers fail to read. - Deploy
fonti_web.pyoralbo.pyto crawl municipal boards and download linked acts into an indexed directory. - Execute
dati_query.pyto perform SQL analysis and numerical integrity checks on extracted CSV or Parquet data.
Frameworks & tools
Python 3.11+, DuckDB, Poppler, Tesseract (Italian language pack), pdfplumber, PyMuPDF, Trafilatura, and yt-dlp.
Why this beats prompting it yourself
Standard LLMs struggle with precise citations, Italian PA portal structures, and large PDF tables. This skill enforces a "citation-first" rule where every output includes the specific page, URL, or timestamp needed for a legal or journalistic dossier.
Use cases
- Auditing municipal procurement by extracting CIG codes and amounts from hundreds of PDF acts.
- Proving a website change by comparing current portal content against historical Wayback Machine snapshots.
- Building a demographic profile of a specific town using direct ISTAT data integration.
Known limitations
Home pages with complex layouts may yield limited text. Demographic coverage for smaller provinces requires manual code entry. Remote table extraction has rate limits.
Home pages may yield limited text. Specific smaller provinces require manual codes for ISTAT data. Remote table extraction is limited to 2 requests per minute. Requires Tesseract and Poppler.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 10 days ago
Requires
Permissions
Allowed Hosts