This server can only be called with an API key (a Smithery proxy address). We cannot connect without one, so no measurements are taken and the metrics below stay empty.

usaspending-mcp-server

https://usaspending-mcp-server--cyanheads.run.tools ↗

usaspending-mcp-server is a remote MCP server published at usaspending-mcp-server--cyanheads.run.tools. Its tool list could not be read; the server is closed to direct calls. It answers direct calls with an authentication wall.

Access US federal award, recipient, agency, and spending analytics data from USAspending.gov.

Auth-gatedWebsite ↗2266 usesVersion: —First probe: —Last probe: —

Agent readability score

Can an LLM agent pick the right tool here — names, descriptions and parameter clarity are assessed.

Agent readability score

A highly sophisticated MCP server providing deep, interconnected access to US federal spending data with exceptional chaining guidance.based on catalog declaration

Strengths

Explicit tool-chaining instructions for complex workflows
Detailed value mappings for codes, types, and recipient levels
Proactive guidance on data preparation requirements like FIPS codes

Weaknesses

Minor text truncation in one tool description
Parameter mandatory/optional status is primarily context-dependent in descriptions

usaspending_disaster_spending — Description text is truncated at the end.

Verdict

This server is exceptionally well-engineered for LLM agents, prioritizing the 'how-to-chain' logic essential for multi-step reasoning. The documentation is precise, significantly reducing the likelihood of parameter errors.

9/4/2026, 4:16:02 PM

Event narrative

Raw observations

Run	Era	Modern	ms	Legacy	ms	Versions

Supported versions

—

as declared by server/discover + initialize

Capabilities

—

not readable

Schema score

—

not measurable

Schema compliance

tools/list structure, inputSchema validity, and a functional smoke test — the components of the 0-100 score.

—

not measurable

tools/list not measurable
inputSchema not measurable
Smoke test not run this round
Cold-start not measurable (no handshake)

Era history

Each cell is one probe run. Faded cells are incomplete probes — one leg did not answer, so the era is inconclusive.

No era measurements yet.

Response time history

The two probe legs separately: modern server/discover and legacy initialize.

No response-time measurements yet.

Event log

Derived by comparing consecutive probes — changes in era, protocol version, build and reachability.

Get your badge

Add this badge to your README — it updates automatically as measurements change.

Markdown

[![mcpmetrics](https://mcpmetrics.io/badge/cyanheads/usaspending-mcp-server/era.svg)](https://mcpmetrics.io/servers/cyanheads-usaspending-mcp-server)

HTML

<a href="https://mcpmetrics.io/servers/cyanheads-usaspending-mcp-server"><img src="https://mcpmetrics.io/badge/cyanheads/usaspending-mcp-server/era.svg" alt="mcpmetrics"></a>

Raw observations

You are seeing the last 7 days. Sign up for the full history. Which check failed and why is in the dashboard.