
Space Mushroom MCP server
Which MCP servers are Space Mushroom's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.
read 2026-11-02
- Registry namecom.spacemushroom/materialsverified in the MCP registry
The registry name sits under spacemushroom.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-11-02
- Tools11
- Server calls itselfspace-mushroom
- Server version0.1.0
- Protocol version2025-06-18
On 2026-11-02 the server answered initialize and tools/list without a sign-in. First answer in 194 ms.
- Its tools do every use case we checked: Compare stainless steel grades on sourced data; Review recorded failure causes for a given material.
- Checked on 2 of the 7 days it takes to judge reliability.
Once connected
- For "Compare stainless steel grades on sourced data", the likely first call is search_materials.
- For "Review recorded failure causes for a given material", the likely first call is get_material.
Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.
- INITE MCP scoreafter a weeka score needs 7 daily checks: 2 so far
- What an agent can do with it2 of 2 use cases
- Answeredmeasuring2 of the 7 daily checks it takes to print a share
- Median first answer1161 ms
- Space in an agent's context≈ 2,224 tokensthe definitions of all its tools (11), which an agent carries on every turn
- Tool definitionsA · 100 of 100
- Typed results0 of 11 toolsdeclared outputSchema: results come typed, read once
- Every tool passes all six checks.
What an agent can do with it
- can do · Compare stainless steel grades on sourced data
- not sure · Check which polymers resist atomic oxygen in orbit
- can do · Review recorded failure causes for a given material
Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?
Worth a look
None of the five warning signs in these definitions.
A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →
search_materialsFind materials by name, abbreviation or formula. Optionally filter by category slug (for example polymers, metals, energy). Returns ids to use with get_material.
get_materialThe sourced record of one material: sourced numbers with conditions, method and source location, plus uses, hazards, failure modes, corrosion and supply facts. Optionally narrow to one property or fact topic.
compareLine several materials up on one property, with each value, its conditions and its source. Values are also given in the property's canonical unit (value_canonical), so they can be ranked directly. Values measured under different…
list_propertiesThe property names in use, with their canonical unit and whether higher or lower is better. Use these names with get_material and compare.
get_termA plain-language definition of a technical term or abbreviation used on material pages, for example ARC, SoC or adiabatic.
find_materialsRequirements in, candidates out. Each material gets a verdict: pass (every constraint met by a sourced value), fail, or unknown (no data on a constraint, so it cannot be ruled in or out). Every constraint reports pass, fail or no_data and…
list_claimsEvery current sourced value for one material (optionally one property) in canonical units, with evidence level and a claim_id. Use the claim_id with prove_claim to see the exact source and passage.
prove_claimEverything behind one value: as printed in the source, converted, the evidence level and what it means, the conditions, and each source with the place in it (and the passage where we have it).
what_is_missingFor one material and a list of properties: which have sourced data (and the strongest evidence) and which have none. "no_data" means not covered yet, not known to be safe.
resolve_designationWhich material a code names: a UNS number, EN number or name, DIN, JIS, AISI or SAE number, ASTM or AMS specification, or trade name (for example S31600, 1.4401, SUS316, A240). Case, spaces and hyphens do not matter, and a short code also…
search_sourcesFull-text search over short passages quoted from our sources. Returns the passage, the source and the place in it. Passages are still being added, so an empty answer means none indexed yet.
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"materials": {
"type": "http",
"url": "https://spacemushroom.com/mcp"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http spacemushroom https://spacemushroom.com/mcpclaude.ai, ChatGPT and other clients: add https://spacemushroom.com/mcp as a custom connector in the settings.
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.