Invokary MCP server

• Registry namecom.invokary/platform-requirementsverified in the MCP registry

The registry name sits under invokary.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

• Stateanswerschecked 2026-10-24
• Tools6
• Server calls itselfplatform-requirements
• Server version1.0.0
• Protocol version2025-06-18

On 2026-10-24 the server answered initialize and tools/list without a sign-in. First answer in 213 ms.

• Its tools do none of the product's use cases we checked: map an existing API to a small set of agent tools; submit a connector to Claude and ChatGPT directories; run evals before each directory submission.
• Checked on 4 of the 7 days it takes to judge reliability.

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: 4 so far
• What an agent can do with it0 of 3 use cases
• Answeredmeasuring4 of the 7 daily checks it takes to print a share
• Median first answer367 ms
• Space in an agent's context≈ 1,064 tokensthe definitions of all its tools (6), which an agent carries on every turn
• Tool definitionsA · 100 of 100
• Typed results6 of 6 toolsdeclared outputSchema: results come typed, read once

• Every tool passes all six checks.

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 →

• list_platforms

  Lists the platforms in the Platform Requirements dataset (Claude, ChatGPT, Cursor and the Official MCP Registry): each one's directory, vendor, number of requirements, topics and latest check date. Use it to see what the dataset covers…

• get_platform_requirements

  Returns what one platform requires before it lists an MCP connector or plugin, each requirement with its sources, evidence label and the date it was last checked. Narrow the list to one topic (category) or one product surface. Platforms…

• get_requirement

  Returns one requirement by its id, for example claude.rule-annotations, with its sources and the date it was last checked. For a retired id it returns the retirement date and the requirement that replaced it.

• compare_platforms

  Sets the requirements of two or more platforms side by side, grouped by topic, to show where they differ, for example on authentication or on what a submission needs. Leave platforms out to compare all four directories.

• search_requirements

  Finds requirements whose text matches keywords or a question, such as "privacy policy", "OAuth" or "test cases", across every platform or within one. Returns up to 10 matches, best first, with their sources and check dates.

• list_platform_changes

  Lists Invokary's dated notes on changes to these platforms' submission rules, most recently updated first, optionally for one platform or updated on or after a date. Each note links to its full text.