swiss-efv-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HOST | No | Bind host (SSE only). Loopback by default; set 0.0.0.0 only in a container | 127.0.0.1 |
| PORT | No | Bind port (SSE only) | 8000 |
| TRANSPORT | No | Transport: stdio (Claude Desktop) or sse / streamable-http (cloud) | stdio |
| EFV_MCP_LOG_LEVEL | No | structlog level (JSON to stderr) | INFO |
| EFV_MCP_CORS_ORIGINS | No | SSE only: explicit allowed browser origins (default-deny; comma-separated or JSON) | [] |
| EFV_MCP_OTEL_ENABLED | No | Enable OpenTelemetry tracing (requires the otel extra); standard OTEL_* env vars configure export | false |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| fiscal_headlineA | Headline fiscal time series: revenue, expenditure, balance and debt ratios
from 1990 to the latest year the EFV publishes, actuals and forward-looking
years alike. Read Use case: track how a federal aggregate evolved over time — e.g. "how did the
Bund balance develop since the 2022 rate turnaround?". variable e.g. 'saldo',
'einnahmen', 'ausgaben', 'bruttoschuldenquote'. household: bund|ktn|gdn|staat|sv.
model: fs|gfs. Every point flags |
| fiscal_budget_breakdownA | Hierarchical federal-budget breakdown for one topic and year. Use case: see where the money goes — e.g. "which task areas absorbed the
spending growth?". topic e.g. 'Ausgaben nach Aufgabengebiet', 'Ausgaben nach
Art', 'Einnahmen'. level is the hierarchy depth (1 = total, 2 = first
breakdown …); 'contains' filters the path substring for drill-down. An empty
result carries a |
| fiscal_by_institutionA | Federal spending by department / administrative unit since 2007. Use case: compare personnel, IT or external-services spending across
departments — e.g. "IT spending of the Finanzdepartement since 2010?".
variable one of: 'Personalausgaben', 'Informatik', 'Beratung und externe
Dienstleistungen', 'Anzahl Vollzeitstellen'. An empty result carries a |
| fiscal_list_dimensionsA | List the valid dimension values across all datasets (variables, households, models, budget topics, departments). Use case: call this first to build correct parameters for the other tools — it turns free-text guesses into exact filter values. Loads all three dumps, so it may take a moment on a cold cache. |
| fiscal_statusA | Report cache freshness and upstream health per dataset. Use case: check whether the data is fresh, cached or degraded before trusting a figure — the health endpoint of this server. Never returns empty silently; used for graceful degradation. |
| dump_statusA | DEPRECATED — use Reports cache freshness and upstream health per dataset (SEC-022: every tool
now shares the |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 6 tools
Each tool targets a distinct aspect of the fiscal data: time series, budget breakdown, institutional spending, dimension discovery, and health status. The only potential overlap (fiscal_status vs. dump_status) is explicitly resolved by deprecating the latter, so agents can clearly distinguish them.
All tools follow the `fiscal_` prefix with a descriptive noun (headline, budget_breakdown, by_institution, list_dimensions, status). The deprecated alias `dump_status` breaks the pattern but is clearly marked as temporary, so the overall convention remains highly consistent.
Six tools cover the core operations for a read-only fiscal data server: querying aggregates, hierarchical breakdowns, institutional comparisons, dimension discovery, and health checks. This is well-scoped without unnecessary bloat or gaps.
The surface covers the primary data access patterns for the domain: time-series aggregates, budget hierarchies, departmental spending, and metadata discovery. The status tool ensures graceful degradation, and the deprecated alias is a minor artifact that does not affect completeness.