Cyan Eclipse Bazaar MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| BAZAAR_CATALOG | No | Override path to the catalog.json file used by the server (catalog lives in catalog.json beside the server by default). |
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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_productsA | Keyword search over the Bazaar catalog (names, capabilities, categories, problem statements). Every hit carries match_reasons — why it matched is never hidden. Optional category filter. Use find_solutions instead when you have a problem to describe rather than keywords; use get_product once you have a product id. |
| find_solutionsA | Describe a problem in plain language; get ranked products with a fit explanation, status, and certification badge inline. Spec-stage products are labelled spec-stage, uncertified products say so. Use search_products instead when you already have keywords or a category in mind; use get_product once you have a product id. |
| get_productA | Full catalog entry for one product id: capabilities, problem statements, pricing model, platforms, tools, and the certification block including what was NOT tested, plus the product's live_state_note and its explicit install_plan status (no install plan exists yet). Use get_certification instead when you only need the bench verdict, scope, and not_tested list for the same id. |
| compare_productsA | Side-by-side comparison of 2+ product ids: status, pricing model, certification verdict and scope, platforms, categories. Use get_product instead when you need one product's full entry. |
| get_certificationA | The latest Test Bench verdict for a product, with a live scorecard summary, the certified scope, and the not_tested list. Returns an explicit NOT YET CERTIFIED answer when no verdict exists. Use get_product instead when you need the full catalog entry for the same id. |
| list_categoriesA | All catalog categories with product counts broken down by status (live / first-party / in-development / spec), plus the product ids in each category. Start here to browse the catalog; use the returned ids with get_product or get_certification, and the category names as search_products' category filter. |
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 has a clearly distinct purpose, and the descriptions actively cross-reference siblings with 'use X instead when' guidance. The only overlaps (search_products vs find_solutions for discovery, get_product vs get_certification for the certification block) are explicitly delineated by input type and output scope.
All six names follow a strict verb_noun snake_case pattern (search_products, get_product, compare_products, get_certification, list_categories, find_solutions). Verbs are chosen precisely for the operation (list for enumeration, get for single retrieval, compare for multi-id).
Six tools is well-scoped for a read-oriented catalog server: three discovery paths, two retrieval paths, and one comparison tool. Nothing appears redundant or padding.
Discovery, retrieval, certification, and comparison are covered, and list_categories supplies ids for browsing. However, there is no tool to fetch or generate the referenced install_plan, no status/platform filtering beyond category, and no write/lifecycle operations if the server is meant to be more than read-only.