aivis-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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 |
|---|---|
| check_ai_visibilityA | Check whether AI assistants can find, read and transact with an ecommerce store. Three checks: (1) robots.txt, distinguishing crawlers whose blocking removes you from AI answers (OAI-SearchBot, PerplexityBot, Claude-SearchBot, Amzn-SearchBot, Applebot) from training-only crawlers where blocking costs nothing (GPTBot, ClaudeBot, Google-Extended); (2) product page RAW HTML — not the rendered DOM, because most AI crawlers do not run JavaScript — covering structured data, offer completeness and how many concrete measurements the page gives; (3) the agent-commerce layer (Universal Commerce Protocol / MCP), which decides whether an agent can actually buy rather than merely describe. Read-only: nothing is created, purchased or uploaded. |
| build_recommendation_testA | Generate the prompt set for testing whether AI assistants actually recommend a brand in its category, plus a scoring rubric. This tool does NOT query any assistant — it cannot, and any tool claiming a definitive 'AI ranking' is ahead of the evidence, because rankings are not public and vary by wording, location and session. What it does is remove the part that does not scale: writing varied, non-leading prompts and scoring the answers consistently. If you (the calling assistant) can search the web, run these yourself and report the results back to the user. Otherwise hand them to the user to run monthly. |
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 2 tools
The two tools target completely different activities: one performs a technical visibility audit (robots.txt, raw HTML, agent-commerce layer), while the other generates prompts and a rubric for recommendation testing. Their descriptions make the distinction unambiguous, so an agent can easily pick the right tool.
Both tool names follow a clear verb_noun pattern in snake_case (check_ai_visibility, build_recommendation_test). The naming is consistent and predictable, with no deviations in style or convention.
The server has only 2 tools, which feels thin even for a focused audit service. While each tool is substantive, the surface lacks depth—for example, no tool to run or score the recommendation test, or to track changes over time—making the set borderline.
The tools cover the core diagnostic tasks: technical AI visibility checks and prompt generation for recommendation testing. There are minor gaps, such as no tool to apply the generated rubric or to store historical results, but these can be worked around by the calling assistant or user.