swiss-procurement-mcp

Overview Schema Source Related Servers Discussions Admin

search_procurements

Read-onlyIdempotent

Search Swiss public procurement projects on simap.ch by topic, canton, CPV code, or date to find open tenders. Returns project IDs for detailed records.

Instructions

Find open tenders matching a topic, canton, CPV code or date window — the default entry point when the question is "what is being tendered?".

Search Swiss public procurement projects on simap.ch.

Covers all cantons and the Confederation, updated intraday. This is the entry point; use get_procurement_details with the returned ids for the full record.

Note that simap indexes PROJECTS, not publications: one hit is one project, represented by its NEWEST publication. A project tendered in March and awarded in July appears once, as the July award. Use get_publication_history to see the earlier publications of a project.

At least one filter is required — simap answers a filterless query with nothing rather than everything.

Args: query: Free-text search over titles and descriptions. canton: Bare canton id, e.g. ZH (NOT CH-ZH). See CANTON_IDS. canton_match: How canton is interpreted. procuring_body (default) — procured by that canton's public bodies, including communal and subordinate offices. place_of_delivery — the work is delivered there. Beware: ~60% of publications carry no structured order address and are invisible to this filter. both — the union of the two. Costs two upstream calls and does not support cursor. cpv_codes: One or more CPV classification codes. Resolve names to codes with search_cpv_codes first. process_type: One of open, selective, invitation, direct, no_process. pub_type: Publication type. For awarded contracts use one of award_tender, award_study_contract, award_competition, direct_award — a plain "award" is rejected by the API. published_from / published_until: ISO dates YYYY-MM-DD. These filter on the NEWEST publication date of a project. cursor: Pagination cursor from a previous response's next_cursor. language: de, fr, it or en.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`args`	Yes

Output Schema

TableJSON Schema

Name	Required	Description	Default
`note`	No
`count`	Yes
`source`	Yes
`results`	Yes
`has_more`	Yes	True if the pagination cursor can be advanced.
`match_type`	No	exact when results were returned, none when empty.	none
`provenance`	Yes
`next_cursor`	No	Pass as `cursor` to search_procurements for the next page.
`retrieved_at`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed4 schema fields changedv0.19.0Sep 27, 2026

addedOutput schema / $defs / LotRef

Added value: +{
+  "description": "One lot of a publication, carrying the id `get_publication_history` needs.\n\nUpstream keeps the publication history per lot: a lot publication answers\n`past-publications` only when a `lotId` comes with it. Dropping this list\nfrom the summary left that id unobtainable through this server, so the\nhistory tool could not be called correctly for any lot-based procurement.",
+  "properties": {
+    "lot_id": {
+      "title": "Lot Id",
+      "type": "string"
+    },
+    "lot_number": {
+      "anyOf": [
+        {
+          "type": "integer"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "default": null,
+      "title": "Lot Number"
+    },
+    "lot_title": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "default": null,
+      "title": "Lot Title"
+    },
+    "pub_type": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "default": null,
+      "title": "Pub Type"
+    },
+    "publication_id": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "default": null,
+      "description": "The lot's own publication id, which may differ from the project's.",
+      "title": "Publication Id"
+    }
+  },
+  "required": [
+    "lot_id"
+  ],
+  "title": "LotRef",
+  "type": "object"
+}

addedOutput schema / $defs / ProcurementSummary / properties / lots

Added value: +{
+  "description": "Empty unless lots_type is 'with'. Pass a lot_id from here to get_publication_history, which cannot trace a lot publication without one.",
+  "items": {
+    "$ref": "#/$defs/LotRef"
+  },
+  "title": "Lots",
+  "type": "array"
+}

addedOutput schema / $defs / ProcurementSummary / properties / lots_dropped

Added value: +{
+  "default": 0,
+  "description": "How many lots upstream listed without a lotId, and this server could therefore not carry through. Above 0 the `lots` list is incomplete: those lots exist upstream, but their history is unreachable from here. Treat a non-zero value as missing data, not as fewer lots.",
+  "title": "Lots Dropped",
+  "type": "integer"
+}

addedOutput schema / $defs / ProcurementSummary / properties / lots_type

Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "with or without — whether the project is split into lots.",
+  "title": "Lots Type"
+}

First observedv0.18.3Jul 31, 2026

TDQS

A5/5.0

Behavior5/5

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

The annotations already mark this as read-only, open-world, idempotent, and non-destructive. Beyond that, the description discloses important behavior: simap indexes projects rather than publications, only the newest publication represents a project, filterless queries return nothing, and canton_match='both' costs two upstream calls and disables pagination.

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?

The description is long but front-loaded and well organized: use case, core behavior, sibling routing, indexing caveat, filter requirement, then Args. Every section adds practical guidance for invoking the tool correctly.

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?

For a complex procurement search tool with an output schema available, the description covers purpose, alternatives, filtering rules, pagination limitations, and regional/classification semantics. Nothing critical for correct invocation appears to be missing.

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

Parameters5/5

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

Although the input schema contains parameter descriptions, the tool description adds substantial semantic guidance: canton must be a bare id like ZH rather than CH-ZH, canton_match meanings and caveats, cpv_codes should be resolved via search_cpv_codes, pub_type 'award' is rejected, dates filter on the newest publication date, and cursor comes from next_cursor.

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?

The description states a specific verb and resource: searching Swiss public procurement projects on simap.ch. It clearly identifies itself as the default entry point for 'what is being tendered?' and distinguishes itself from get_procurement_details and get_publication_history by explaining the project/publication indexing difference.

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

Usage Guidelines5/5

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

It explicitly says when to use this tool ('default entry point when the question is what is being tendered?'), how to follow up with get_procurement_details and get_publication_history, and how to resolve CPV codes with search_cpv_codes. It also states the non-obvious rule that at least one filter is required.

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

Deploy Server