This server acts as an MCP server for Oblio.eu, enabling natural language interaction (e.g., via Claude Desktop) to manage fiscal documents, payments, e-Factura submissions, and reference data.
Document Management
Create invoices (factura), proformas, or delivery notices (aviz) with full line item details; supports an idempotencyKey to prevent duplicate issuance on retries
Get a single document by series and number
List documents with filters (series, client, date range, status, pagination)
Cancel a document (reversible)
Restore a previously cancelled document
Delete a document permanently (only the last in a series)
Payments
Record a payment (incasare) against an existing invoice, specifying method (e.g., Chitanta, Card, Ordin de plata), amount, and date; defaults to the full remaining balance
e-Factura / SPV (ANAF)
Submit issued invoices to Romania's SPV (e-Factura / ANAF)
Check SPV processing status (0 = processing, 1 = success, 2 = errors, -1 = not sent)
Reference Data (Nomenclatures)
Query companies, saved clients, products, VAT rates, document series, languages, and management units (gestiuni) to ensure valid inputs for document creation
Webhooks
Subscribe external endpoints to Oblio events (stock changes, document lifecycle updates, payment recordings)
List and delete webhook subscriptions
Multi-company Support
Switch the active company CIF at runtime and check which CIF is currently in use, allowing management of multiple companies from a single Oblio account
Configuration
API credentials (email, secret, optional default CIF) are set via environment variables, with optional OAuth token persistence
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@oblio-mcp-server Issue invoice to client CIF RO12345678 for 1500 RON"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol server for Oblio.eu — issue invoices, proformas and delivery notices, collect payments, submit e-Factura to Romania's SPV, and query your account's reference data through natural language, from Claude Desktop or any MCP client.
Unofficial community project. Not affiliated with or endorsed by Oblio Software; it builds on their public API and official Node SDK.
Built on Oblio's official SDK (@obliosoftware/oblioapi) — no hand-rolled HTTP layer for the core API
Full document lifecycle — create, fetch, list, cancel, restore and delete invoices, proformas and delivery notices (avize)
Payments — record collections (incasari) against invoices, with all Oblio payment methods
e-Factura / SPV — submit invoices to ANAF's SPV and poll their processing status
Nomenclatures — discover valid series names, VAT rates, saved clients, products, languages and management units before issuing documents
Webhooks — subscribe external endpoints to Oblio events (stock changes, document drafts/updates/cancellations, recorded payments)
Multi-company — switch the active company CIF at runtime via set_cif
Client-side idempotency guard — pass an idempotencyKey to create_document and retries won't double-issue a document
Zod-validated inputs — every tool input is schema-checked with descriptive field docs
MCP tool annotations — read-only / destructive / idempotent hints on every tool, so clients can gate confirmations appropriately
Related MCP server: docnova-mcp
Node.js >= 20
An Oblio.eu account with API access
In Oblio, go to Setari > Date Cont. You need:
the account email (acts as the OAuth client id)
the API secret (acts as the OAuth client secret)
your company CIF (e.g. RO12345678) — optional at startup; it can also be set at runtime with the set_cif tool
Add to your MCP client configuration (e.g. claude_desktop_config.json). Straight from GitHub (npx builds it on first run):
{
"mcpServers": {
"oblio": {
"command": "npx",
"args": ["-y", "github:horatiuvlad/oblio-mcp-server"],
"env": {
"OBLIO_API_EMAIL": "you@example.com",
"OBLIO_API_SECRET": "your-api-secret",
"OBLIO_CIF": "RO12345678"
}
}
}
}Once the package is published to npm, "args": ["-y", "oblio-mcp-server"] will work as well.
git clone https://github.com/horatiuvlad/oblio-mcp-server.git
cd oblio-mcp-server
npm install
npm run build{
"mcpServers": {
"oblio": {
"command": "node",
"args": ["/path/to/oblio-mcp-server/dist/index.js"],
"env": {
"OBLIO_API_EMAIL": "you@example.com",
"OBLIO_API_SECRET": "your-api-secret",
"OBLIO_CIF": "RO12345678"
}
}
}
}Variable | Required | Description |
| yes | Oblio account email (Setari > Date Cont) |
| yes | Oblio API secret (Setari > Date Cont) |
| no | Default company CIF; can be changed at runtime with |
| no | Path where the OAuth access token is persisted so it survives restarts (defaults to in-memory only) |
Tool | Description |
| Issue an invoice, proforma or delivery notice (aviz); supports an optional |
| Fetch a single document by series name and number, with totals, status and a link |
| List documents with filters: series, number, client (cif/email/phone/code), issue-date range, draft/cancelled/collected flags, optional line items / payments / SPV status, sorting and pagination |
| Cancel (annul) a document; it stays in Oblio and can be restored later |
| Restore a previously cancelled document to its active state |
| Permanently delete a document (Oblio only allows deleting the last one in a series) |
Tool | Description |
| Record a payment (incasare) against an existing invoice; defaults to the full remaining amount when no value is given |
Tool | Description |
| Fetch reference data: companies, saved clients, products, VAT rates, document series, languages, or management units (gestiuni) |
Tool | Description |
| Submit an issued invoice to Romania's SPV (e-Factura / ANAF) |
| Fetch the SPV status and archive for a submitted invoice (0 = processing, 1 = success, 2 = errors, -1 = not sent) |
Tool | Description |
| Subscribe an endpoint to an Oblio event: |
| List all webhook subscriptions with topic, endpoint and id |
| Delete a webhook subscription by id |
Tool | Description |
| Set the active company CIF used for subsequent requests (switch between companies on one account) |
| Return the company CIF currently in effect |
It sits on Oblio's official SDK rather than reimplementing the API surface
It ships with a test suite and strict typechecking
create_document has a client-side idempotency guard, so agent retries don't silently issue duplicate invoices
Full e-Factura / SPV coverage — submit and track invoices with ANAF, not just issue them locally
npm run build # compile to dist/
npm test # run the test suite
npm run typecheck # tsc --noEmitOblio publishes no machine-readable API spec — the HTML docs at
oblio.eu/api are the de-facto contract that
src/schemas.ts hand-encodes. Three layers guard against silent drift:
Docs snapshot — scripts/oblio-docs-snapshot.mjs extracts the
contract-bearing parts of the docs page (endpoints, parameter tables,
response samples) into docs/oblio-api.snapshot.md. The
oblio-docs-drift workflow
re-fetches weekly and, on any change, opens a PR with the refreshed
snapshot and the diff.
Agent draft — when the ANTHROPIC_API_KEY repo secret is configured,
the same workflow runs Claude Code on the drift PR branch to translate the
docs diff into real src/schemas.ts / tool changes (including
.describe() text), verified with typecheck + tests and pushed as a
commit on the PR. A human still reviews and merges; without the secret the
PR is opened with a manual checklist instead.
SDK tripwire — Renovate flags releases of
@obliosoftware/oblioapi, which usually accompany API changes.
node scripts/oblio-docs-snapshot.mjs --check # exit 3 + diff on drift
node scripts/oblio-docs-snapshot.mjs --update # refresh the snapshotCancel (annul) a document. The document stays in Oblio marked as cancelled and can later be restored with restore_document.
| Name | Required | Description | Default |
|---|---|---|---|
| type | Yes | Document type: "invoice" (factura), "proforma", or "notice" (aviz). | |
| number | Yes | Document number. | |
| seriesName | Yes | Document series name. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Adds context beyond annotations: specifies document remains in system and can be restored. Clearly describes non-destructive nature despite destructiveHint, giving accurate behavioral insight.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, no redundant words, front-loaded with purpose. Every sentence adds value.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Explains the outcome and restoration possibility. Lacks mention of prerequisites (e.g., document must be in cancellable state), but is otherwise sufficient for a simple cancellation tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema covers all 3 parameters with descriptions (100% coverage). Description adds no additional parameter information, so baseline score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states verb 'cancel/annul' and resource 'document', explains effect (marked as cancelled, restorable), and distinguishes from sibling tools like delete_document and restore_document.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Mentions restoration capability and names restore_document as alternative, but could be more explicit about when to use vs delete_document (e.g., reversible vs permanent deletion).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Record a payment (incasare) against an existing invoice, identified by series name and number. The collect object specifies the payment method (Chitanta, Ordin de plata, Card, etc.) and optionally the amount and date; when no value is given the full remaining invoice amount is collected. Applies to invoices only.
| Name | Required | Description | Default |
|---|---|---|---|
| number | Yes | Invoice number. | |
| collect | Yes | A payment collection. | |
| seriesName | Yes | Invoice series name. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations indicate it's a write operation (destroy hint false, read-only false). Description adds that it records a payment, consistent with annotations. No behavioral surprises.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three well-structured sentences. Front-loaded with main purpose, then details on collect object, ending with scope limitation. No wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Covers purpose, key parameters, and default behaviors. No output schema, so return values are not explained, but that's acceptable. Sufficient for a payment recording tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, but description adds meaningful defaults: when amount is omitted, full remaining invoice is collected; date defaults to today. Also explains seriesName requirement for Chitanta. Adds value beyond schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Specifically states it records a payment against an invoice identified by series and number. Distinguishes from sibling tools like create_document or cancel_document by focusing on payment collection.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Provides context that payment applies only to invoices and explains default behavior for amount and date. However, no explicit when-not-to-use or alternative tool names given.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Issue a new fiscal document in Oblio: an invoice (factura), a proforma, or a delivery notice (aviz). Provide the document body (client, series, line items, dates). NOTE: Oblio has no draft state — the document is issued immediately. Pass an idempotencyKey to guard against accidental double-issue on retries.
| Name | Required | Description | Default |
|---|---|---|---|
| data | Yes | Document body. | |
| type | Yes | Document type: "invoice" (factura), "proforma", or "notice" (aviz). | |
| idempotencyKey | No | Optional caller-supplied key. If the same key was already used in this server process, the previous result is returned instead of issuing a second document. Guards against accidental double-issue on retries. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds value beyond the annotations by disclosing that Oblio has no draft state (document is issued immediately) and advising use of an idempotencyKey to prevent double-issue. The annotations indicate readOnlyHint=false and openWorldHint=true, which are consistent. However, it does not detail return format or error conditions, so a score of 4 is fitting.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise with three sentences: first states the purpose, second outlines required inputs, third highlights critical behavioral notes. No unnecessary words. Front-loaded with the action and document types. Excellent structure.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of the input schema (nested objects, many fields) and lack of output schema, the description is relatively brief. It covers the core purpose and key caveats (no drafts, idempotency) but omits return value expectations, error handling, prerequisites (e.g., company configuration), and how to use referenceDocument. The schema itself is well-documented, but the description could be more complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the baseline is 3. The description summarizes the key components ('client, series, line items, dates') but does not add significant meaning beyond what the schema already provides. The idempotencyKey mention is helpful but within the description. The score remains 3.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Issue a new fiscal document') and lists the specific document types (invoice, proforma, notice). It is not a tautology, but it does not explicitly differentiate from the sibling tool 'create_einvoice', which may be confusing. Overall, the purpose is clear and specific.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for creating fiscal documents in Oblio, but it does not provide explicit when-to-use or when-not-to-use guidance compared to alternatives like 'create_einvoice'. The note about no draft state and idempotency is behavioral rather than usage context. The guidance is minimal, so a score of 3 is appropriate.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Submit an existing Oblio invoice to Romania's SPV (e-Factura). The invoice must already be issued in Oblio; this sends its XML to ANAF. The response includes an einvoice status: 0 = processing, 1 = success, 2 = errors, -1 = not sent.
| Name | Required | Description | Default |
|---|---|---|---|
| number | Yes | Invoice number. | |
| seriesName | Yes | Invoice series name to submit to SPV. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations indicate a write operation (readOnlyHint=false) and non-idempotent behavior. The description adds valuable behavioral details by listing possible response statuses (0=processing, 1=success, 2=errors, -1=not sent), which helps the agent understand outcomes. No contradictions with annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two concise sentences cover purpose, prerequisite, and response status. Every sentence adds value with no redundant or extraneous information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a two-parameter tool with no output schema, the description adequately covers the action, prerequisites, and response status. It does not detail all possible error scenarios, but the provided information is sufficient for an agent to use the tool effectively.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with descriptions for both parameters ('Invoice number.' and 'Invoice series name to submit to SPV'). The description does not add new meaning beyond the schema, so it meets the baseline but does not exceed it.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('submit'), the resource ('existing Oblio invoice to Romania's SPV'), and the context ('e-Factura'). It distinguishes from siblings like 'get_einvoice' (retrieval) and 'create_document' (creation) by specifying that the invoice must already exist in Oblio.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides a clear prerequisite ('invoice must already be issued in Oblio') and mentions that it sends XML to ANAF. However, it does not explicitly state when not to use the tool or mention alternatives, though the context is strong enough for an agent to infer appropriate usage.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Permanently delete a document. Oblio only allows deleting the last document in a series; prefer cancel_document for anything else.
| Name | Required | Description | Default |
|---|---|---|---|
| type | Yes | Document type: "invoice" (factura), "proforma", or "notice" (aviz). | |
| number | Yes | Document number. Only the last in a series can be deleted. | |
| seriesName | Yes | Document series name. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations show destructiveHint=true; description adds permanence and series constraint, enhancing beyond annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, no waste, front-loaded with action and constraint.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Covers purpose, condition, and alternative; lacks output description but sufficient for a delete tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema covers all parameters with descriptions; description adds 'only last in series' already in number parameter description, no extra value.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Clearly states 'Permanently delete a document', distinguishing from sibling 'cancel_document'.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicitly specifies condition (only last in series) and alternative (prefer cancel_document).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Return the company CIF currently used for Oblio requests.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false. The description adds the context 'used for Oblio requests' but does not disclose additional behavioral traits beyond what annotations provide.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence of 10 words, directly stating the purpose without any extraneous information. It is perfectly concise and front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description is nearly complete for a simple read tool with no parameters and rich annotations. It lacks only a mention of the return format, but without an output schema this is a minor gap.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters, so schema coverage is complete (100%). The description adds no parameter information, which is acceptable as there are none. Baseline for 0 parameters is 4.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb 'Return' and clearly identifies the resource as 'the company CIF currently used for Oblio requests'. It distinguishes from the sibling tool 'set_cif' by focusing on retrieval.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
While the name and description imply retrieval, there is no explicit guidance on when to use this tool versus alternatives like 'set_cif'. The context is clear but lacks explicit when-not criteria.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fetch a single document (invoice/proforma/notice) by its series name and number, including totals, status and a link.
| Name | Required | Description | Default |
|---|---|---|---|
| type | Yes | Document type: "invoice" (factura), "proforma", or "notice" (aviz). | |
| number | Yes | Document number within the series. | |
| seriesName | Yes | Document series name. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, so the safety profile is clear. The description adds that the tool returns totals, status, and a link, but does not disclose any behavioral traits beyond this (e.g., rate limits, error conditions).
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is one sentence, front-loaded with the purpose, and every element is necessary. No redundancy or filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 3 required parameters and no output schema, the description adequately explains what the tool returns (totals, status, link). However, it could be more specific about the format of totals or status fields. Annotations cover safety, so overall it is fairly complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with each parameter already described (type enum, number, seriesName). The description adds no additional meaning beyond the schema, just references the parameters. Baseline score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'Fetch' and the resource 'single document (invoice/proforma/notice)', specifying the identifying parameters (series name and number) and included data (totals, status, link). This distinguishes it from sibling tools like list_documents and cancel_document.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description does not explicitly state when to use this tool versus alternatives (e.g., get_einvoice, list_documents). While the purpose is clear, there is no guidance on prerequisites or scenarios where other tools would be more appropriate.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fetch the SPV (e-Factura) status and archive for an invoice previously submitted with create_einvoice. Status codes: 0 = processing, 1 = success, 2 = errors, -1 = not sent.
| Name | Required | Description | Default |
|---|---|---|---|
| number | Yes | Invoice number. | |
| seriesName | Yes | Invoice series name to submit to SPV. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, destructiveHint=false. Description adds value by listing status codes (0,1,2,-1) and their meanings, which aids in understanding response behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two concise sentences, front-loaded with action and resource, no extraneous information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
No output schema, but description explains return includes status and archive, plus status codes. Could elaborate on archive content, but sufficient for a fetch operation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%. Description provides context that parameters refer to the original invoice submitted with create_einvoice, but schema descriptions already cover the basic meaning. Adds modest value.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states verb 'Fetch', resource 'SPV status and archive', and context 'for an invoice previously submitted with create_einvoice'. Distinguishes from siblings by specifying prerequisite tool.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Indicates usage after create_einvoice, but lacks explicit when-not-to-use or alternative tools. Provides sufficient context for correct usage.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fetch Oblio reference data: companies on the account, saved clients, products, VAT rates, document series, languages, or management units (gestiuni). Use this to discover valid values (e.g. series names, VAT rate names) before creating documents. Supports an optional name filter and extra query filters such as pagination offset.
| Name | Required | Description | Default |
|---|---|---|---|
| name | No | Optional name filter (clients/products). | |
| type | Yes | Reference data set to fetch. | |
| filters | No | Extra query filters, e.g. { offset: 250, cif: 'RO123' }. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already indicate readOnlyHint, idempotentHint, and destructiveHint. The description adds useful behavioral context by specifying the types of data fetched and the ability to apply name and extra query filters, which goes beyond the annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is three sentences long, front-loaded with purpose, and each sentence adds unique value: purpose, usage scenario, and additional capabilities. No unnecessary words or fluff.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a tool with 3 parameters, no output schema, and nested objects, the description adequately covers the purpose, usage, and filtering options. It could mention return format, but it's not essential given the tool's nature.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% and descriptions in the schema are clear. The description adds value by recontextualizing the 'name' parameter as 'optional name filter' and giving an example for 'filters' (pagination offset), reinforcing the schema details.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states 'Fetch Oblio reference data' and lists specific data sets (companies, clients, products, etc.). The verb 'Fetch' and resource 'reference data' are specific, and the tool is distinct from sibling tools that involve creating, cancelling, or deleting documents.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly advises 'Use this to discover valid values ... before creating documents,' providing a clear use case. It also mentions optional filters like pagination, but does not explicitly state when not to use the tool, though the context is clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List documents of a given type, with optional filters: series, number, client, issue-date range, draft/cancelled flags, sorting and pagination.
| Name | Required | Description | Default |
|---|---|---|---|
| id | No | Filter by Oblio document id. | |
| type | Yes | Document type: "invoice" (factura), "proforma", or "notice" (aviz). | |
| draft | No | Draft filter: -1 all, 0 issued only, 1 drafts only. | |
| client | No | Filter by client identity (any combination of cif/email/phone/code). | |
| number | No | Filter by document number. | |
| offset | No | Pagination offset. | |
| orderBy | No | Sort field. | |
| canceled | No | Cancelled filter: -1 all, 0 active only, 1 cancelled only. | |
| orderDir | No | Sort direction. | |
| collected | No | Payment filter: -1 all, 0 uncollected, 1 collected. | |
| seriesName | No | Filter by series. | |
| issuedAfter | No | Issue date lower bound YYYY-MM-DD. | |
| issuedBefore | No | Issue date upper bound YYYY-MM-DD. | |
| limitPerPage | No | Results per page (max 100). | |
| withCollects | No | 1 to include payment collections. | |
| withProducts | No | 1 to include line items in the response. | |
| withEinvoiceStatus | No | 1 to include e-Factura (SPV) status. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, openWorldHint, idempotentHint true, and destructiveHint false, covering safety. The description adds the filter options but no additional behavioral traits beyond what annotations provide, so the bar is met but not exceeded.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence of 21 words, front-loaded with the core purpose 'List documents of a given type'. Every part (filters, pagination) is essential and adds value, with zero fluff.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a list tool with 17 parameters, rich schema, and annotations covering safety, the description adequately summarizes the filter capabilities and mentions pagination. It doesn't detail return structure but the schema covers pagination parameters, and annotations compensate for behavioral aspects.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all 17 parameters. The description summarizes the filter categories (series, number, client, etc.) but doesn't add meaning beyond the schema, earning the baseline score of 3.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'List' and resource 'documents' with a required type, and lists specific filters (series, number, client, date range, draft/cancelled, sorting, pagination). This distinguishes it from siblings like get_document (single) or create_document (creation).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for listing documents by type with optional filters, providing clear context for when to use this tool. However, it does not explicitly mention when not to use it or alternative tools, though the sibling list makes that implicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Restore a previously cancelled document back to its active state.
| Name | Required | Description | Default |
|---|---|---|---|
| type | Yes | Document type: "invoice" (factura), "proforma", or "notice" (aviz). | |
| number | Yes | Document number. | |
| seriesName | Yes | Document series name. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations indicate the tool is not read-only (writes), is idempotent, and not destructive. The description adds the critical context that it only works on cancelled documents, which is valuable beyond annotations. No contradictions.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, concise sentence of 10 words. It conveys the essential information without any wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (3 parameters, no output schema, no nested objects), the description is fully complete. It states the purpose, pre-condition (cancelled document), and result.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the schema already documents all parameters well. The description does not add additional meaning or examples beyond what the schema provides.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('restore'), the resource ('previously cancelled document'), and the outcome ('back to its active state'). It distinguishes from the sibling 'cancel_document' tool.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies the tool is for cancelled documents but does not explicitly state when to use it versus alternatives like 'create_document' or 'cancel_document'. No 'when not to use' guidance is given.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Set the active company CIF used for all subsequent Oblio requests. Use this to switch between companies on the same Oblio account.
| Name | Required | Description | Default |
|---|---|---|---|
| cif | Yes | Company CIF to use for subsequent requests, e.g. RO45079498. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations indicate this is a write operation (readOnlyHint=false) that is idempotent and non-destructive. Description adds critical context that it affects all subsequent requests, which goes beyond annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Description is concise with two short sentences. It is front-loaded with the purpose and uses no unnecessary words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple tool with one parameter and no output schema, the description together with annotations provides complete context. It addresses the state-changing nature and persistence across requests.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Only one parameter 'cif' with a clear schema description and example. Since schema coverage is 100%, the description adds no extra meaning beyond what the schema provides, meeting the baseline.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Title and description clearly state the tool sets the active company CIF for subsequent Oblio requests. It is distinct from sibling 'get_cif' which retrieves the current CIF, and other siblings are for different operations.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Description explicitly says to use this tool to switch between companies on the same account. It does not explicitly state when not to use, but the context is clear given siblings like 'get_cif'.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0cancel_documentcollect_paymentcreate_documentcreate_einvoicedelete_documentget_cifget_documentget_einvoiceget_nomenclatureslist_documentsrestore_documentset_cifScored across 12 tools
Each tool has a clearly distinct purpose: document lifecycle (create, get, list, cancel, restore, delete, collect payment), e-invoice (create, get status), company configuration (get/set CIF, get nomenclatures). No overlapping functionality that would confuse an agent.
All tool names follow a consistent verb_noun pattern in lowercase_snake_case (e.g., cancel_document, create_einvoice, list_documents). No mixing of conventions or confusing abbreviations.
With 12 tools, the server is well-scoped for a document management system. Each tool covers a necessary operation without redundancy or excess. The count is appropriate for the domain.
The tool set covers the core lifecycle of fiscal documents: create, read, list, cancel, restore, delete, and payment collection, plus e-invoice integration. A minor gap is the lack of an update tool, but this is typical for fiscal documents that cannot be modified after issuance. Overall sufficiently complete.
Invoicing you drive by talking to your AI: log time, raise invoices and track what's owed via MCP.
Connect Exact Online to your AI assistant via MCP. Manage Exact Online with natural language.
AI-native ERP MCP: ES/EU fiscal compliance (VeriFactu/TicketBAI/Facturae), invoicing, tax, banking
Fakturownia.pl (Polish invoicing) via MCP: invoices, clients, products, warehouse. Full read-write.
