The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Swiss Procurement MCP listing page.
Part of the Swiss Public Data MCP Portfolio — open-source MCP servers connecting AI agents to Swiss public and open data.
This is a private project. It is independent of any employer or institutional affiliation and represents no official position of any authority.
MCP server for Swiss public procurement — read access to the official simap.ch API, covering all cantons and the Confederation, updated intraday.
«Which school-building tenders did the City of Zurich publish in 2026, which BKP construction categories do they concern, and who are the procuring offices?»
A single search_procurements_detailed(query="Schulhaus", canton="ZH", published_from="2026-01-01")
returns the leading tenders already expanded with their BKP construction codes and
procuring offices — connecting procurement to school-building planning in one call
(optionally paired with search_construction_codes to resolve a category).
Swiss public procurement is published on simap.ch. The platform's web UI is
searchable by hand, but the amtsblatt-mcp
server only reaches the three cantons (AR, BS, TI) that still mirror tenders to
the Amtsblattportal — Zurich among the missing.
simap closes that gap: it operates a documented OpenAPI 3 read API (v1.5.1)
whose search and detail endpoints are marked security: None and are callable
without authentication. This server wraps exactly those read endpoints.
Mnemonic: The web UI is the front door; the API is the loading dock. Probe the dock.
Architecture A (live API only, short-lived cache).
my/ / OIDC-protected endpoints (publishing tenders,
submitting offers) are out of scope — this server never writes.Every response carries source and provenance (live_api / cached /
degraded). Upstream failure yields a degraded envelope, never a silent empty
list.
| Endpoint | Auth | Result |
|---|---|---|
/publications/v2/project/project-search | none | 20 hits, canton filter, current-day |
/publications/v1/.../publication-details/... | none | full record: criteria, deadlines, codes |
/publications/v1/publication/{id}/past-publications | none | project lifecycle |
/codes/v1/cpv/search | none | CPV full-text search |
/codes/v1/{bkp,npk,ebkp-h,ebkp-t,oag,cpc}/search | none | Swiss construction codes |
/procoffices/v1/po/public | none | ~1 MB office list (client-side filter) |
/cantons/v1, /countries/v1 | none | reference data |
www.simap.ch/api.
The simap.ch/de web UI is a separate SSR app that exposes none of it —
probing the UI produced an earlier, mistaken "no API" verdict.lang is mandatory on project-search. Omitting it is HTTP 400
(errorCode E0025), not an empty result. The client injects a default.newestPubTypes=award returns HTTP 400. Awards are
split by procedure: award_tender, award_study_contract,
award_competition, direct_award. The search_awards tool queries all four.ZH, not CH-ZH. Passing an ISO subdivision code
silently matches nothing; this server rejects it with a clear error.| Tool | Purpose |
|---|---|
search_procurements | Search projects by canton, CPV, process type, date, text |
search_procurements_detailed | Search + full detail for the top n hits in one call (aggregated) |
search_awards | Awarded contracts only (all four award types at once) |
get_procurement_details | Full record for one publication |
get_publication_history | Earlier publications of the same project (tender → award) |
search_cpv_codes | Resolve keywords to CPV classification codes |
search_construction_codes | Swiss construction codes (BKP, NPK, eBKP, OAG, CPC) |
find_procurement_office | Public procurement offices by partial name |
source_status | Reachability and latency of the simap.ch API |
All tools carry readOnlyHint, idempotentHint and openWorldHint (they query
the live simap.ch API).
Every tool takes a single validated argument object. Bounds, allow-lists and
patterns are declared on the input models in
inputs.py — so an out-of-range limit or
an unknown canton is rejected before any upstream request, and the constraints
are visible to the model in the tool schema rather than buried in the tool body:
The models set strict=True (no silent "10" → 10 coercion) and
extra="forbid" (unknown fields are rejected, not ignored). The canton, process
type, publication type, code system and language allow-lists are derived from
constants.py, so they cannot drift from the probe-verified tables.
canton= meanssimap offers exactly one geographic filter, orderAddressCantons, and it selects
by where the work is delivered — not by who is procuring. When a procuring
office files a free-text address, the structured canton is null and the
publication is invisible to that filter. Measured CH-wide over 500 projects
published since 2026-07-01: 303 (60.6%) carry no canton, among them the Amt
für Hochbauten Zürich, Grün Stadt Zürich, USZ, BBL and SBB.
canton_match therefore makes the question explicit:
| Value | Matches | Zurich, 2026-07-01…27 |
|---|---|---|
procuring_body (default) | procured by that canton's public bodies, incl. communal and subordinate offices (issuedByOrganizations) | 410 projects |
place_of_delivery | the work is delivered there (orderAddressCantons) | 263 projects |
both | union of the two; two upstream calls, no pagination | 441 projects |
The 31 projects only place_of_delivery finds are federal bodies procuring in
Zurich (ETH, Empa, Flughafen Zürich AG) — a different question, not a gap, which
is why this is three explicit semantics rather than a silent union.
Every response states in note which semantics were applied.
register-mcp.zh-education-mcp.amtsblatt-mcp with
national coverage instead of three cantons.The image is multi-stage and runs as a non-root system user. compose.yaml
adds a read-only root filesystem, drops all capabilities, sets
no-new-privileges, and caps memory, CPU and PIDs. No secret is needed at
runtime — the wrapped simap.ch endpoints are public.
CI builds the image on every push and asserts both properties that matter:
that the container does not run as uid 0, and that the server still imports
under --read-only --cap-drop ALL.
| Variable | Default | Purpose |
|---|---|---|
MCP_TRANSPORT | stdio | stdio | sse | streamable-http |
MCP_HOST / HOST | 127.0.0.1 | HTTP binding (cloud transports only). Defaults to loopback; set 0.0.0.0 explicitly to expose all interfaces in a cloud deployment. |
MCP_CORS_ORIGINS | (unset) | Comma-separated origins allowed to call the HTTP transports from a browser. Unset means no cross-origin browser access at all — stdio and non-browser clients are unaffected. Mcp-Session-Id is exposed and accepted for the listed origins, so a browser client can hold a session. * is honoured but logs a warning and disables credentials, because browsers reject a wildcard origin together with credentials. |
MCP_STATELESS | (off) | Set to 1 to run the streamable-http transport with no session tracking. Removes session affinity as a concern for multi-instance deployments; gives up SSE stream resumption and server-initiated notifications. No effect on the legacy SSE transport, which logs a warning if asked. See docs/load-balancing.md. |
PORT / MCP_PORT | 8000 | HTTP port (cloud transports only) |
LOG_LEVEL | INFO | DEBUG | INFO | WARNING | ERROR. Structured JSON, one object per line, always on stderr — stdout carries the MCP protocol on a stdio transport. |
No API keys — the wrapped simap.ch read endpoints are fully public.
Built on structlog. Every event emitted during a
tool call carries that call's correlation_id, bound via contextvars — so a
failure logged deep inside the HTTP client can be joined to the request that
caused it without threading context through every function.
| Level | Emitted when |
|---|---|
DEBUG | a tool call was entered (tool_call_started) — tells you whether a hung call ever started |
INFO | a tool call finished cleanly, with latency |
WARNING | simap.ch was unreachable or errored (upstream_degraded) |
ERROR | a tool call raised |
Records carry the exception type only — never its message and never an upstream response body (OBS-002).
Served via the initialize handshake | 2024-11-05 … 2025-11-25 — the handshake ceiling |
| Served via the per-request envelope | 2026-07-28 |
| Who picks | The client's first request, once per connection. A request carrying the 2026-07-28 _meta envelope opens a modern connection; anything else opens a handshake connection. |
| Pinned in | MCP_PROTOCOL_VERSION in server.py |
| SDK | mcp>=2.0.0,<3 |
| Cache hints | tools/list and server/discover: ttlMs 300000, cacheScope public |
The MCP Python SDK negotiates the protocol version in the session layer and offers no constructor parameter for it, so the version cannot be pinned by configuration. It is pinned as a declared constant and enforced by detection:
protocol_version_drift event at WARNING. The server keeps working.tests/test_protocol_version.py fails.That split is deliberate. An SDK bump should break our build, not the runtime
of someone who upgraded mcp in their own environment.
.github/dependabot.yml).CHANGELOG.md in one commit.CHANGELOG.md, not folded
into a dependency-bump line.This server exposes tools and neither resources nor prompts. That is a decision, not an omission, so here is the reasoning (ARCH-008).
Why not resources. Resources address identifiable, listable content —
GET-like reads the client can enumerate and cache. simap's endpoints are the
opposite: every useful call is a query with filters over a corpus of ~200k
publications that changes intraday. A resource URI would either enumerate
something unbounded or encode a full query in the URI, which is a tool with
extra steps.
Two tools were checked concretely for migration potential and rejected for specific reasons, not by blanket policy:
| Candidate | Why it stays a tool |
|---|---|
source_status | Genuinely resource-shaped — one fixed, cacheable document. But it exists to be called when a result looks wrong, and a resource the model has to remember to re-read is worse at that job than a tool it can invoke on suspicion. |
search_cpv_codes | The CPV catalogue is finite and stable enough to enumerate. But it is ~10k entries; exposing it as a resource would push the whole classification into the context window, when the point of the tool is that the server does the lookup. |
Why not prompts. A curated prompt list would encode question templates ("which tenders in canton X…"). The tool docstrings already carry that guidance where the model actually reads it, and prompts would duplicate it in a second place that can drift — this repo has already been bitten twice by exactly that class of duplication.
This will be revisited if the server ever gains a genuinely enumerable, slow-changing dataset.
See EXAMPLES.md for use cases grouped by audience (schools, public, administration, developers) and a tool-selection reference table.
project-search indexes projects and
represents each by its newest publication. A project tendered in March and
awarded in July appears once, as the July award; search_awards likewise only
finds projects whose newest publication is an award, so a later correction
hides it. get_publication_history reaches the earlier publications.get_publication_history needs a lot_id whenever the
search result shows lots_type: "with" — take one from that result's lots
list. Without it the source answers HTTP 400 and the tool reports a degraded
response naming the missing parameter. Measured 2026-08-29 over 80
publications: all 4 with lots behaved this way, all 76 without lots answered
directly.tools/ packageThe portfolio structure standard asks for a tools/ package once a server
exposes more than five tools. This one exposes nine and keeps them in
server.py, which is a deliberate deviation rather than an oversight — recorded
here because this is where the standard, and anyone comparing against it, looks.
server.py is ~900 lines and the surrounding modules above are already split out
by concern, so the intent of the standard — a codebase navigable without
scrolling one omnibus file — is met. What the split would add is the literal file
layout.
The companion server amtsblatt-mcp is the case where it was worth doing: its
server.py had grown to 2477 lines holding HTTP plumbing, XML parsing, a
taxonomy cache, the input models and every handler, and it was split in that
project's 0.21.0. That refactor is also the reason for caution here — it
introduced a defect (an extracted module captured a cache global by value, so a
tool silently reported stale state) that the entire test suite passed
through, because no test covered the affected path. It was caught by reading
the diff.
Moving nine handlers for the literal form of a standard whose intent is already
satisfied would take that risk for no navigational gain. This should be revisited
if server.py passes roughly 1500 lines.
Phase 1 — read-only (see ROADMAP.md for the phase-specific backlog and what a phase transition would require). This server wraps only the public read endpoints; the write / OIDC-protected simap endpoints are deliberately out of scope. See the SECURITY.md re-evaluation triggers for the conditions that would move it to a write phase.
The server targets the MCP spec version pinned as MCP_PROTOCOL_VERSION — see
MCP Protocol Version above for the current value and
how the pin is enforced. SDK and dependency updates arrive as
Dependabot PRs, so a breaking protocol or SDK change
is reviewed deliberately rather than drifting in silently.
Contributions are welcome — see CONTRIBUTING.md for how to report bugs, suggest a new endpoint, or submit code.
This is a read-only, no-PII, public-open-data server. Audited against the
portfolio MCP best-practice catalogue (15 pass / 16 partial / 1 fail across
32 applicable checks, production-ready). See SECURITY.md for the
posture and how to report a vulnerability, and audits/ for the full
report.
MIT License — see LICENSE. The tenders are official public-procurement announcements; simap.ch publishes no explicit open-data licence, so reuse follows the simap.ch terms (see Credits).
Hayal Oezkan · github.com/malkreide
See CHANGELOG.md.
/api/specifications/simap.yaml, which a live test checks the enum constants against. Guides: kissimap.ch.mcp-data-source-probe methodology.The code in this repository is MIT licensed; the data is simap.ch's, under its terms (see above). Public money, public code.