WirStartenKI

AI Impact Analysis

ai_impact_analysis

Where AI pays off first in a company. 15 multiple-choice questions: role, size, industry, revenue, biggest cost block, where the time goes, AI use today, data, ownership, know-how, repetitive share, headcount, standardisation, mandate, timing. Returns maturity, impact and momentum scores (0 to 100) with readings, three ranked AI levers, a conservative effect range for the cost block and a short assessment. Deterministic and free; no sign-in, the e-mail is optional.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "role": "<role>",
      "size": "<size>",
      "industry": "<industry>",
      "revenue": "<revenue>",
      "cost_block": "<cost_block>",
      "time_sink": "<time_sink>",
      "tools": "<tools>",
      "data": "<data>",
      "governance": "<governance>",
      "skills": "<skills>",
      "repetitive_share": "<repetitive_share>",
      "headcount": "<headcount>",
      "standardisation": "<standardisation>",
      "priority": "<priority>",
      "timeline": "<timeline>"
    },
    "name": "ai_impact_analysis"
  }
}

Input Schema

Name	Required	Description	Default
`data`	Yes	Where does the data for that block live? silos = In e-mails and Excel; mixed = Partly in ERP, CRM or PLM; central = Mostly in one system; apis = Structured and accessible via interfaces
`role`	Yes	What is your role? founder_ceo = Founder or CEO; cxo = CxO or managing director; head = Head of a division or department; manager = Team lead or project lead; other = Something else
`size`	Yes	How many people work in your company? s50 = Under 50; s250 = 50 to 250; s1000 = 250 to 1,000; s1000plus = More than 1,000
`email`	No	Optional. With an e-mail the result is saved and the reply carries a personal report link.
`tools`	Yes	How is AI used in your company today? none = Not at all; individuals = A few people use ChatGPT and the like; pilots = First pilot projects; rollout = Live in one area; embedded = Part of several processes
`skills`	Yes	How much AI know-how is in the team? none = Hardly any; enthusiasts = A few enthusiasts; trained = Trained users; experts = People who build with it
`company`	No
`revenue`	Yes	Roughly, what is your annual revenue? r5 = Under 5 million; r25 = 5 to 25 million; r100 = 25 to 100 million; r500 = 100 to 500 million; r500plus = More than 500 million; none = Rather not say
`industry`	Yes	Which industry are you in? automotive = Automotive; industrial = Industrial manufacturing; energy = Energy; hightech = Electronics and high-tech; appliances = Consumer appliances; other = Another industry
`language`	No	Language of the result	de
`priority`	Yes	Is there a mandate or budget for it? none = Not yet; listed = It is on the list; this_year = Planned for this year; budget = Budget is approved
`timeline`	Yes	When do you want to start? none = Just exploring; m12 = Within twelve months; m6 = Within six months; now = Now
`headcount`	Yes	How many people work in that block? lt5 = Under 5; h5_20 = 5 to 20; h20_100 = 20 to 100; gt100 = More than 100
`time_sink`	Yes	Where does most of the time go in that block? documentation = Writing documents and reports; reporting = Collecting figures and reporting; search = Searching for information; customer_comms = Answering customers and partners; data_entry = Moving data between systems; coordination = Meetings and coordination
`cost_block`	Yes	Where is your biggest cost block? engineering = Engineering and development; sales_marketing = Sales and marketing; service = Service and after-sales; operations = Production, purchasing and supply chain; finance_admin = Finance and administration; it_data = IT and data
`first_name`	No
`governance`	Yes	Who owns the AI topic? none = Nobody yet; discussed = It is discussed, nobody owns it; policy = There are rules, but no owner; owner = One person or team with a mandate
`standardisation`	Yes	How standardised is the work there? adhoc = Everyone does it their own way; partly = Partly, with templates; documented = Documented processes; measured = Documented and measured
`repetitive_share`	Yes	How much of the work in that block is repetitive? lt10 = Under 10 percent; s10_30 = 10 to 30 percent; s30_50 = 30 to 50 percent; gt50 = More than half

Schema Changelog

Changes observed during successful MCP inspections.

AddedNov 8, 2026

TDQS

A3.7/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations declare readOnlyHint=false and idempotentHint=false, leaving the 'what gets created' question open; the description fills that gap by saying the e-mail optionally saves the result and returns a personal report link. It also adds that the computation is deterministic and requires no sign-in, which is real behavioral context beyond the annotations. Response shape is sketched but not fully specified.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, front-loaded with the payoff statement, then the input topics, then the returned artifacts. Dense but every clause carries information for a 19-parameter tool; the parenthetical '(0 to 100)' is the only slightly filler-like fragment.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 19-param tool with no output schema, the description covers the return payload (maturity, impact, momentum scores, three ranked levers, effect range, short assessment) and the side effect of supplying an e-mail. What remains unclear is the response format and whether the questionnaire is answered in one call, but the essentials for correct invocation are present.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 89% with per-enum value explanations, so the schema already does the semantic work. The description's topic list roughly mirrors the 15 required fields and flags the e-mail as optional, but adds no value-level or ordering meaning beyond what the schema states. Baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The opening sentence plus the enumeration of the 15 question topics and the returned scores makes it unmistakable that this is a questionnaire-driven AI maturity/impact assessment. It is easily separable from about_wirstartenki, book_call and get_open_slots. It loses a point only because no explicit verb ('submit the assessment and receive...') frames the action.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies usage by describing inputs and outputs, and notes 'Deterministic and free; no sign-in, the e-mail is optional', which lowers the barrier to calling it. However there is no explicit when-to-use, no statement of what must already be known before calling, and no routing relative to siblings such as book_call or get_open_slots.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All WirStartenKI tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?