GCC e-Invoice Preflight Oracle
tech.noahsarch/einvoice-preflightnoahsarch.techstreamable-httpv0.1.0Scanned
Validate UAE PINT AE and Saudi ZATCA e-invoices: UBL, official Schematron, QR, fix hints.
7 tools: 5 change things and 2 only read.
- Reads private data · 0
- Reads text others write · 0
- Can send data out · 0
- No findings
Anlyon alternative
5 of these tools change something. Run those as governed actions.
Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.
Anlyon does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.
Import an OpenAPI spec →Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
7 tools
Changes things 5Claude Code: ask
check_trnFormat check of a UAE Tax Registration Number (15 digits) and derivation of the TIN / Peppol participant identifier (scheme 0235 = first 10 digits).No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Format check of a UAE Tax Registration Number (15 digits) and derivation of the TIN / Peppol participant identifier (scheme 0235 = first 10 digits). Free. Example: trn '100 1234 5670 0003'. Response: result with verdict (ACCEPTED | ACCEPTED_WITH_WARNINGS | REJECTED), findings[] (id, severity, message, fields, location, fix_hint, source = xsd | schematron | preflight), counts, ruleset version and timings, plus data_confidence, estimated_items[], sources[], data_version, warnings[] and a disclaimer. Documents are validated in memory and never stored or logged.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- readOnlyHint: true
extract_summarySeller/buyer (names, TRN, endpoint, legal registration, emirate), dates, currency, totals, VAT breakdown, line count, payment means, attachments and the detected rule set — without…No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Seller/buyer (names, TRN, endpoint, legal registration, emirate), dates, currency, totals, VAT breakdown, line count, payment means, attachments and the detected rule set — without validating. Free. Response: result with verdict (ACCEPTED | ACCEPTED_WITH_WARNINGS | REJECTED), findings[] (id, severity, message, fields, location, fix_hint, source = xsd | schematron | preflight), counts, ruleset version and timings, plus data_confidence, estimated_items[], sources[], data_version, warnings[] and a disclaimer. Documents are validated in memory and never stored or logged.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- readOnlyHint: true
readiness_reportBatch preflight for a go-live check or an ERP regression suite: per-document verdicts, readiness percentage, the most frequent rejection rules across the batch, optional printable …No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Batch preflight for a go-live check or an ERP regression suite: per-document verdicts, readiness percentage, the most frequent rejection rules across the batch, optional printable HTML. Charged per document. Response: result with verdict (ACCEPTED | ACCEPTED_WITH_WARNINGS | REJECTED), findings[] (id, severity, message, fields, location, fix_hint, source = xsd | schematron | preflight), counts, ruleset version and timings, plus data_confidence, estimated_items[], sources[], data_version, warnings[] and a disclaimer. Documents are validated in memory and never stored or logged.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- readOnlyHint: true
zatca_qr_decodeDecode the base64 TLV of a Saudi e-invoice QR into seller name, VAT number, timestamp, totals, hash, signature and key tags, with phase detection and format issues.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Decode the base64 TLV of a Saudi e-invoice QR into seller name, VAT number, timestamp, totals, hash, signature and key tags, with phase detection and format issues. Free. Response: result with verdict (ACCEPTED | ACCEPTED_WITH_WARNINGS | REJECTED), findings[] (id, severity, message, fields, location, fix_hint, source = xsd | schematron | preflight), counts, ruleset version and timings, plus data_confidence, estimated_items[], sources[], data_version, warnings[] and a disclaimer. Documents are validated in memory and never stored or logged.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- readOnlyHint: true
zatca_qr_encodeEncode the five phase-1 fields (plus optional phase-2 hash/signature/key tags) into the base64 TLV string that goes into the QR image and the UBL QR reference.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Encode the five phase-1 fields (plus optional phase-2 hash/signature/key tags) into the base64 TLV string that goes into the QR image and the UBL QR reference. Free. Response: result with verdict (ACCEPTED | ACCEPTED_WITH_WARNINGS | REJECTED), findings[] (id, severity, message, fields, location, fix_hint, source = xsd | schematron | preflight), counts, ruleset version and timings, plus data_confidence, estimated_items[], sources[], data_version, warnings[] and a disclaimer. Documents are validated in memory and never stored or logged.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- readOnlyHint: true
Only reads 2Show 2 read-only tools
list_rulesRule ids, severities and texts from the installed PINT AE Schematron (shared PINT rules and the -ae jurisdiction rules), filterable by severity or text, with fix hints where we hav…No findings
Why it is in this group
- The name "list_rules" says "list": a word for reading.
Description, as the server wrote it Rule ids, severities and texts from the installed PINT AE Schematron (shared PINT rules and the -ae jurisdiction rules), filterable by severity or text, with fix hints where we have them. Free. Example: contains 'TaxPointDate'. Response: result with verdict (ACCEPTED | ACCEPTED_WITH_WARNINGS | REJECTED), findings[] (id, severity, message, fields, location, fix_hint, source = xsd | schematron | preflight), counts, ruleset version and timings, plus data_confidence, estimated_items[], sources[], data_version, warnings[] and a disclaimer. Documents are validated in memory and never stored or logged.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- readOnlyHint: true
validate_invoiceFull preflight of one UBL Invoice or CreditNote: XML well-formedness, UBL 2.1 XSD, the official PINT AE Schematron (≈300 rules incl.No findings
Why it is in this group
- The name "validate_invoice" says "validate": a word for reading.
Description, as the server wrote it Full preflight of one UBL Invoice or CreditNote: XML well-formedness, UBL 2.1 XSD, the official PINT AE Schematron (≈300 rules incl. the UAE jurisdiction overlay) or the ZATCA pre-check, plus advisory cross-checks. Returns a verdict and every finding with rule id, severity, fields, location and fix hint. Example: xml '<Invoice …>', jurisdiction 'auto'. Response: result with verdict (ACCEPTED | ACCEPTED_WITH_WARNINGS | REJECTED), findings[] (id, severity, message, fields, location, fix_hint, source = xsd | schematron | preflight), counts, ruleset version and timings, plus data_confidence, estimated_items[], sources[], data_version, warnings[] and a disclaimer. Documents are validated in memory and never stored or logged.
Annotations the server declares
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
- readOnlyHint: true
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or untrusted-input or outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write0
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 0.1.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http einvoice-preflight https://einvoice.noahsarch.tech/mcpclaude mcp add --transport http einvoice-preflight https://einvoice.noahsarch.tech/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__einvoice-preflight__validate_invoice",
"mcp__einvoice-preflight__list_rules"
],
"ask": [
"mcp__einvoice-preflight__readiness_report",
"mcp__einvoice-preflight__extract_summary",
"mcp__einvoice-preflight__check_trn",
"mcp__einvoice-preflight__zatca_qr_decode",
"mcp__einvoice-preflight__zatca_qr_encode"
],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as einvoice-preflight. Review it before you use it.
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://anlyon.dev/api/v1/servers/tech.noahsarch/einvoice-preflight/verify?hash=62432edec3fff285c056b4cbb40927cb93a76c5826c9c90a0c713b9c55b90774" | jq -e '.changed == false'Pinned hash 62432edec3ff, from Oct 31, 2026, 21:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://anlyon.dev/servers/tech.noahsarch/einvoice-preflight/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://anlyon.dev/servers/tech.noahsarch/einvoice-preflight)Source: MCP Registry. Synced Oct 31, 2026, 21:18 UTC.