Search the Foundry44 library and install animated UI components for any stack.
Scored across 2 tools
search_components is purely for discovery and returns names, while get_component is purely for retrieval/implementation of a named component. Their purposes are clearly distinct and the descriptions make the hand-off relationship explicit.
Both names follow a consistent verb_noun snake_case pattern (get_component, search_components). The only minor deviation is singular vs plural noun usage between the two.
Two tools is thin for a component library server; discovery and retrieval are covered but there is no room for categories, listing, or metadata operations. Still, for a narrow read-only library surface it is borderline acceptable.
The search-then-get workflow forms a complete path from finding a component to implementing it, with options and pitfalls documented in get_component. Minor gaps exist around browsing/listing the whole catalog or checking component metadata without fetching full implementation steps.
Steps to add a Foundry44 component to the user's project, for any stack (plain HTML, React with or without shadcn, other), with its markup, options and pitfalls. Follow them in the user's project.
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes | The component's name, as search_components gives it (e.g. 'spotlight-card'). | |
| settings | No | Optional values that differ from the defaults: an option by its name (e.g. {"duration": 800}); a colour by its CSS variable (e.g. {"--f44-spotlight-card-color": "secondary"}) with a palette role of the project (brand, secondary, text, background, muted) or a #rrggbb hex. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations exist, so the description must carry the behavioral load. It usefully signals that the tool returns instructions rather than mutating the project ('Follow them in the user's project') and previews the return content (markup, options, pitfalls). It says nothing about read-only guarantees, permission/auth needs, or failure behavior for an unknown component name.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences: a scope/content statement followed by a direct action instruction, front-loaded with the deliverable. 'To the user's project' and 'in the user's project' repeat slightly, costing a little tightness.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a two-parameter tool with a fully described schema the description covers the essentials, and there is no output schema to explain. However, with no annotations it should say more about whether the tool touches the filesystem and what the response looks like structurally (a single step list? per-stack sections?).
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the name and settings parameters are already fully documented, including the settings key/value format and palette roles. The description adds one useful implicit point — that no stack parameter is required because the steps cover every stack — but nothing about settings beyond the schema. Baseline 3 applies.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a concrete deliverable: installation steps for a Foundry44 component including markup, options and pitfalls. That distinguishes it from a naive reading of 'get_component' as a source fetcher. It never names search_components, so sibling differentiation is left to the schema text.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
'For any stack (plain HTML, React with or without shadcn, other)' plus 'Follow them in the user's project' implies the context of use: applying a component to the user's own codebase. There is no explicit when-not, no mention of the search_components prerequisite, and no statement about what to do if the name is unknown.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Find animated UI components in the Foundry44 library by what they do (e.g. 'text reveal on scroll', 'cursor glow'). Returns names to pass to get_component.
| Name | Required | Description | Default |
|---|---|---|---|
| query | No | Words describing the effect, or a component's name. | |
| category | No | Only this family. | |
| freeOnly | No | Only free (MIT) components, the ones that can be installed today. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full disclosure burden; it does establish that this is a lookup whose result is names rather than full component payloads, which is genuinely useful. However, it says nothing about result counts, ranking/relevance behavior, whether an empty result is possible, or any auth or rate constraints, leaving real gaps for an unannotated tool.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences, front-loaded with the purpose and examples, ending with the downstream handoff. Every clause earns its place and nothing is repeated from the schema.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Three optional parameters, all fully documented in the schema, and no output schema means the description must say something about returns — which it does ('Returns names to pass to get_component'). That covers what an agent needs to call it and act on the result; only the absence of result-volume or ranking expectations keeps it from a 5.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the baseline is 3. The description implies intent-based querying and mentions the free-vs-paid distinction implicitly through the library framing, but adds no syntax, format, or combination semantics for query, category, or freeOnly beyond what the schema already documents.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Specific verb ('Find') plus a scoped resource ('animated UI components in the Foundry44 library') and the query style it accepts ('by what they do'). It also names the sibling handoff ('names to pass to get_component'), so an agent can separate search from fetch without opening either schema.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description establishes the search-then-fetch workflow by stating the output feeds get_component, and the two examples ('text reveal on scroll', 'cursor glow') show the intended query style. It stops short of stating when not to use this tool (e.g. when a component name is already known and get_component should be called directly), so it misses explicit exclusions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
get_componentsearch_componentsSearch and install interactive React components with unusual motion, via the shadcn CLI.
Browse and install Crucible's animated React components into shadcn-style projects.
Search, read and install agent-ready React components, screens and blocks for AI products.
Search, read and add Frontic UI's Vue components for commerce storefronts.
Add one secure layer between your agents and this server.
