Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the API NotaGuard, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fnotaguard.com.br%2Fnotaguard" | shStep 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fnotaguard.com.br%2Fnotaguard" | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
What an agent can do with API NotaGuard API.
List NF-e, NFC-e, and NFS-e documents filtered by tipo, status, vinculo, cfop, valor_min, and date ranges
Open one fiscal document by UUID or access key and add taxes, references, tags, or installments with incluir
Read a document's line items with per-item CFOP and the SEFAZ events recorded against it
Download the authorised XML and render the DANFE or DANFSe PDF for a document
GET STARTED
Get aggregated totals for a filtered set of documents to close a period
Sync incrementally by passing the last atualizada_em value to data_atualizacao_apos
Create a batch ZIP export of XML and PDF files and poll its status
Patterns agents use API NotaGuard API for, with concrete tasks.
★ AI bookkeeping agent for received invoices
An AI bookkeeping agent needs the invoices a Brazilian company received so it can match them against payables. Through Jentic the agent searches for 'list received fiscal documents', gets the NotaGuard list operation with its input schema, and calls it with vinculo=receptor and a data_ref_de to data_ref_ate period. It opens each document with incluir=tributos,duplicatas to read taxes and payment installments, with the bearer key injected at call time.
Call GET /v1/notas with tipo=NFe, vinculo=receptor, data_ref_de=2026-07-01, and data_ref_ate=2026-07-31, then call GET /v1/notas/{id} with incluir=tributos,duplicatas for each result and total the installments
Incremental ERP sync
An ERP integration keeps a local copy of fiscal documents without rescanning the whole archive. It stores the highest atualizada_em it has received and passes that instant to data_atualizacao_apos on the next run, paging with proximo_cursor until tem_mais is false. Filtering by vinculo=receptor returns the received archive, which arrives with complete XML.
Call GET /v1/notas with vinculo=receptor and data_atualizacao_apos=2026-11-09T00:00:00Z, follow proximo_cursor until tem_mais is false, and record the highest atualizada_em for the next run
Period close with totals and batch export
At month end, an accountant needs aggregated values and the document files for the period. GET /v1/notas/resumo returns totals for the same filters as the listing, and POST /v1/notas/exportacoes creates a ZIP export with the required formatos field set to xml, pdf, or both. The default arquivo delivery runs asynchronously and is tracked through GET /v1/notas/exportacoes/{id}.
Call GET /v1/notas/resumo with tipo=NFe and the month's data_ref_de and data_ref_ate range, then call POST /v1/notas/exportacoes with formatos set to xml and pdf and the same filtros, and poll GET /v1/notas/exportacoes/{id} until it is done
Document audit by CFOP and SEFAZ events
A tax reviewer wants every document carrying specific CFOPs and its event history. The cfop filter on GET /v1/notas matches documents with any item in the listed codes, GET /v1/notas/{id}/itens shows the CFOP per item, and GET /v1/notas/{id}/eventos lists the SEFAZ events such as manifestations recorded on the document.
Call GET /v1/notas with cfop=5905,6905, then call GET /v1/notas/{id}/itens and GET /v1/notas/{id}/eventos for each document and list items with their CFOP and the latest event
12 endpoints — the notaguard api reads the brazilian fiscal documents that notaguard monitors for a company: nf-e, nfc-e, and nfs-e.
METHOD
PATH
DESCRIPTION
/v1/notas
List fiscal documents
/v1/notas/{id}
Get one fiscal document
/v1/notas/resumo
Get aggregated totals
/v1/notas/{id}/itens
List document items
/v1/notas/{id}/eventos
List SEFAZ events
/v1/notas/{id}/xml
Download the authorised XML
/v1/notas/{id}/pdf
Render the DANFE or DANFSe PDF
/v1/notas/exportacoes
Create a batch export
/v1/notas
List fiscal documents
/v1/notas/{id}
Get one fiscal document
/v1/notas/resumo
Get aggregated totals
/v1/notas/{id}/itens
List document items
/v1/notas/{id}/eventos
List SEFAZ events
/v1/notas/{id}/xml
Download the authorised XML
/v1/notas/{id}/pdf
Render the DANFE or DANFSe PDF
/v1/notas/exportacoes
Create a batch export
What agents get from Jentic-routed access to this vendor.
Setup
Wiring NotaGuard by hand means sending a per-company bearer key, paging with proximo_cursor, handling Brasilia-time date filters, and polling asynchronous exports. Through Jentic you install once, import NotaGuard from the API Directory, store the key once, and your agent calls it.
Permission scoping
NotaGuard puts the document id in the URL path (/v1/notas/{id}/xml, /v1/notas/{id}/pdf), so a rule can bound your agent to specific read operations, and since each key opens one company the credential itself limits which company's documents are reachable. You choose the operations it may call, so the batch export POST is not included unless you add it.
Credential isolation
Your NotaGuard API key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'list received fiscal documents' or 'download an invoice XML', and Jentic returns the matching NotaGuard operation with its input schema, including filters like vinculo, cfop, and data_ref_de, so the agent calls the right endpoint directly.
Alternatives and complements available in the Jentic catalogue.
Specific to using API NotaGuard API through Jentic.
What authentication does the NotaGuard API use?
Per its OpenAPI spec, the NotaGuard API uses HTTP bearer authentication with a company API key in the ng_live_ format, sent in the Authorization header. Keys are created in the NotaGuard panel under Configurações, Integrações, API, and each key opens the documents of a single company. Through Jentic the key is held by your own instance and injected at call time.
Can I download the XML and DANFE PDF with the NotaGuard API?
Yes. GET /v1/notas/{id}/xml returns the authorised XML and GET /v1/notas/{id}/pdf renders the DANFE or DANFSe. The {id} can be the UUID from the listing or the document's access key. Documents that only have a summary still return a summary XML, but the PDF needs the complete XML.
What are the rate limits for the NotaGuard API?
The spec states a usage limit of 60 cost units per minute, per key, and GET /v1/me reports the remaining limit in the current window. Optional incluir blocks cost only the reads you ask for. Check the NotaGuard documentation at https://notaguard.com.br for current details.
Why does a NotaGuard document show status resumida without a value?
For a received document, SEFAZ delivers a summary first and the full document after the recipient registers a manifestation such as ciência da operação, so the state is usually transient. For a document your company issued, SEFAZ does not return the XML to the issuer, so it stays a summary until you upload the XML in the panel. Filtering with vinculo=receptor returns the received archive.
How do I list received invoices for a month through Jentic?
Search Jentic for 'list received fiscal documents', load the NotaGuard list operation, and call GET /v1/notas with vinculo=receptor and data_ref_de and data_ref_ate set to the month. Follow proximo_cursor until tem_mais is false. To run it on your own infrastructure, install Jentic One from its GitHub repo.
Is there a NotaGuard MCP server?
You do not need an MCP server to give your agent NotaGuard. Jentic connects it directly from the API Directory: import it, store your API key once, and your agent calls the document, item, and file operations it needs on demand.
Can I limit what my agent is allowed to do with the NotaGuard API?
Yes. Write a rule that allows GET /v1/notas, GET /v1/notas/{id}, and GET /v1/notas/{id}/xml, and the agent can find documents and fetch their XML but cannot start batch exports through POST /v1/notas/exportacoes. The calls it makes are logged for review.
For Agents
List and filter a company's NF-e, NFC-e, and NFS-e fiscal documents, read their items, taxes, and SEFAZ events, and download the authorised XML or DANFE PDF.
Use for: List every NF-e my company received last month, Get the totals of authorised invoices for July, Download the XML for an invoice by its access key, Retrieve the DANFE PDF for a received NF-e
Not supported: Does not issue, cancel, or register manifestations on fiscal documents with SEFAZ, and does not calculate tax. Use for reading and exporting a company's monitored NF-e, NFC-e, and NFS-e documents only.
The NotaGuard API reads the Brazilian fiscal documents that NotaGuard monitors for a company: NF-e, NFC-e, and NFS-e. A client can list and filter documents by type, status, the company's role as issuer or recipient, CFOP, value range, and reference or issue dates in Brasilia time, then open one document with optional blocks for taxes, references, tags, payment installments, and CFOPs. It returns line items and SEFAZ events, downloads the authorised XML, and renders the DANFE or DANFSe PDF. The API is read-only apart from one operation, which creates an asynchronous or direct batch export of XML and PDF files as a ZIP.
This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.
Base layer of spec validity and structural soundness.
Aggregated quality score from linter diagnostics, weighted by severity.
Percentage of `$ref` references that resolve successfully.
Checks whether the API description parses successfully and conforms to its declared specification (e.g., OpenAPI).
Structural correctness score based on schema issues using logarithmic dampening.
Clarity, completeness, and ingestion readiness for developers and tooling.
How richly the API is illustrated with examples.
Percentage of examples that conform to their schemas.
Percentage of operations with complete response definitions (success, client error, server error).
Health of API ingestion, bundling, and resolution within Jentic pipelines.
Semantic breadth, depth, and agent comprehension for AI systems.
Coverage of descriptions across API elements.
Coverage of RFC 9457 Problem Details for error responses.
Coverage, uniqueness, and casing consistency of operationIds for AI inference.
Coverage of summaries across operations/tags/info.
Functional utility, complexity comfort, and AI orchestration readiness.
Agent comfort level based on API operational and structural complexity.
Trust, risk posture, and security compliance.
Average quality of security schemes based on authentication method strength (weakest link for OAuth2).
Findability, semantic richness, and reasoning readiness.
Clarity and depth of descriptions across API elements.
Score it yourself
Every API in the directory is allowlisted, so you can re-score it with no key required.
npx @jentic/api-scorecard-cli score <openapi-url>