Leadgen MCP

Watch a company for changes

watch_company

Idempotent

Register a company by CUI to baseline its ONRC registry record, then track monthly changes in directors, CAEN activity, address, status, and ANAF VAT/inactive flags.

Instructions

Register a company (by CUI) to monitor. Stores a fingerprint of its current registry record as the baseline; poll_watchlist then reports changes between monthly ONRC snapshots - a new director, a new CAEN activity, an address or status change - and when ANAF declares it fiscally inactive or its VAT status flips.

Input Schema

Name	Required	Description	Default
`cui`	Yes	Romanian tax identification number (CUI) of the firm to watch, e.g. '2816464'. Watching a firm you already watch changes nothing.

Output Schema

Name	Required	Description
`ok`	No	true when the watch is in place.
`asOf`	No	What to cite: registry (ONRC snapshot date), financialYear (fiscal year of the figures), anaf (day of the ANAF data), snapshot, published, source.
`error`	No	Set only when there is no answer (empty_query, invalid_cui, data_unavailable, plan_required ...); such a call is not counted.
`watch`	No	kind, target, owner (key prefix), county, baseline.
`message`	No	A sentence for the caller.
`attribution`	No	The credit line to keep when redistributing the data (CC BY 4.0).
`alreadyWatching`	No	true when you already held this watch (nothing changed).

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv0.1.3Oct 19, 2026

addedOutput schema / properties / attribution

Added value: +{
+  "description": "The credit line to keep when redistributing the data (CC BY 4.0).",
+  "type": [
+    "string",
+    "null"
+  ]
+}

First observedv0.1.0Oct 17, 2026

TDQS

A4.5/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Adds substantial behavior beyond the annotations: it stores a fingerprint of the current registry record as a baseline, changes are computed between monthly ONRC snapshots, and it enumerates the change classes detected (director, CAEN activity, address, status, ANAF fiscal-inactivity, VAT flip). It also reinforces idempotency ('watching a firm you already watch changes nothing'), consistent with idempotentHint=true.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

A single dense sentence, front-loaded with the action, then the baseline mechanism, then the reported outcomes and cadence. No filler sentences and nothing repeated from the schema.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With one required parameter, full schema coverage, an output schema, and annotations covering the safety profile, the description supplies everything else an agent needs: what registration stores, when results appear (monthly snapshots), and via which sibling.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% and the schema already defines CUI with an example plus the idempotency note. The description only restates 'by CUI', adding no format or constraint detail beyond the schema, so the baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a concrete verb+resource ('Register a company ... to monitor') and immediately names the downstream consumer (poll_watchlist), which separates it from siblings like watch_caen and unwatch. An agent can pick it without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explains the workflow context clearly: register here, then poll_watchlist reports changes. It does not state exclusions or explicitly contrast with watch_caen (CAEN-specific watching) or unwatch, so it stops short of full when/when-not guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server