Hyperiux MCP Server
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HYPERIUX_DEBUG | No | Set to 1 to enable verbose request logs — do not use in shared CI. | |
| HYPERIUX_TOKEN | No | Use a CLI token without saving it locally (useful in CI). | |
| HYPERIUX_API_URL | No | Override the API URL independently of the app URL. | |
| HYPERIUX_APP_URL | No | Override the Hyperiux app URL for self-hosting or testing. | |
| HYPERIUX_REGISTRY_URL | No | Override the registry URL for local development. |
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
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| hyperiux_list_effectsA | Browse or search the Hyperiux Vault catalog of React/Next.js interaction effects (scroll systems, cursor trails, WebGL scenes, animated buttons, page transitions, etc). Each result includes tier ("free" or "pro") - use this to answer "how many free/pro effects are there" or "which of these are free" directly from one call, by filtering/counting the returned list yourself. Don't estimate or guess these counts, and don't call hyperiux_get_effect per-item just to total them up - it's unnecessary here. hyperiux_get_effect is still the right call for one specific effect's full description/props/changelog. Args:
Returns JSON: { total, count, offset, effects: [{ name, category, categories, dependencies, version, tier }], has_more, next_offset? } Examples:
|
| hyperiux_get_effectA | Get full details for one Hyperiux Vault effect by its exact slug - description, tier (free/pro), npm dependencies, version, changelog, and install instructions. Optionally includes full source code. Args:
Returns JSON: { name, title, description, tier, version, dependencies, changelog, install_command, preview_url, import_path, target, main, import_statement, files: [{ path, content? }] }
No token (or an invalid one) with include_source=true: the response still includes metadata, but files have no content and sign_in_required/sign_in_required_reason tell the user to run Pro effects without a Pro subscription: metadata only, and source_locked explains the effect needs Hyperiux Pro - expected, not an error. Free effects past the caller's daily install cap: the response still includes metadata but files have no content, and rate_limited/rate_limit_reason explain the daily limit and when to retry - also not an error, it's expected once the cap is hit. Examples:
Error Handling:
|
| hyperiux_list_categoriesA | List every effect category in the Hyperiux Vault catalog with how many effects are in each - useful before calling hyperiux_list_effects with a category filter, to confirm the exact category slug and see relative size. Returns JSON: { categories: [{ category, count }] }, sorted by count descending. Examples:
|
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 3 tools
Each tool targets a distinct operation: browse/search the catalog, fetch one effect's full detail, and enumerate category slugs with counts. Descriptions go further than most servers by including explicit 'Don't use when' guidance pointing at the sibling tool to use instead, eliminating the list-vs-get overlap.
All three tools follow a strict `hyperiux_verb_noun` pattern (list_effects, get_effect, list_categories) with a uniform server prefix. No mixing of camelCase, vague verbs, or divergent conventions.
Three tools is on the lean side of the 3-15 sweet spot, but each earns its place for a read-only catalog browser: discovery, drill-down, and taxonomy. Nothing feels padded or redundant.
List + get + categories covers catalog discovery and detail well, including source retrieval, install metadata, and clear error/rate-limit paths. Minor gaps: no tier or dependency filter (agents must over-fetch and count client-side, as the description itself admits) and no dedicated install/setup tool, leaving that to the CLI.