Overview

analyze provides analytics and reporting capabilities for the KanseiLink ecosystem. It can quantify token savings from using KanseiLink, audit agent costs across services, and generate AEO (Agent Experience Optimization) reports and publishable articles.

This is primarily an admin/operator tool used for business intelligence and content generation.

Modes

analyze supports 4 modes. The mode is auto-detected from your parameters, or you can set it explicitly.

token_savings

Quantify how many tokens agents save by using KanseiLink's pre-verified data versus trial-and-error with raw API docs.

services (string[]): List of service IDs to analyze (e.g., ["freee", "kintone"]). Omit to analyze the top 10 most-used services.

task (string): Optional task context (e.g., "create invoice in freee") to tailor the analysis.

Example

analyze({
  "mode": "token_savings",
  "services": ["freee", "smarthr", "kintone"]
})

analyze({
  "task": "create invoice in freee"
})

cost

Audit agent costs across services. Returns cost breakdowns and optimization recommendations sorted by impact.

cost_service_id (string): Audit a specific service, or omit for all services. Triggers cost mode.

period_days (integer, default: 30): Analysis period in days. Default: 30.

top_n (integer, default: 10): Max recommendations to return (default: 10, max: 50). Sorted by priority (high first) then by monthly_savings_usd descending.

min_priority (enum, default: "low"): Minimum priority level to include: "high" returns only impactful recommendations, "low" returns everything. Default: "low".

Example

analyze({
  "cost_service_id": "freee",
  "period_days": 90,
  "min_priority": "high"
})

analyze({
  "mode": "cost",
  "top_n": 5,
  "min_priority": "medium"
})

aeo_report

Generate an AEO (Agent Experience Optimization) report with scores and rankings for services. Used for consulting deliverables and vendor feedback.

aeo_service_id (string): Filter by service ID. Triggers aeo_report mode when present (without article-specific params).

category (string): Filter by category (e.g., "accounting", "hr", "crm"). When used alone (without aeo_service_id), triggers token_savings mode instead.

aeo_top_n (number, default: 20): Number of top services to return. Default: 20.

include_recommendations (boolean, default: true): Include improvement recommendations per service. Default: true.

Example

analyze({
  "aeo_service_id": "freee",
  "include_recommendations": true
})

analyze({
  "mode": "aeo_report",
  "category": "accounting",
  "aeo_top_n": 10
})

aeo_article

Generate a publishable AEO ranking article with category deep-dives. Outputs in markdown or JSON format.

article_type (string): Article type hint. Triggers aeo_article mode when present.

target_keyword (string): Target keyword for the article. Triggers aeo_article mode when present.

quarter (string, default: "Q2 2026"): Report period label (e.g., "Q2 2026", "2026年上半期"). Default: "Q2 2026".

format (enum, default: "markdown"): Output format: "markdown" for blog/press releases, "json" for API embedding. Default: "markdown".

article_top_n (number, default: 20): Number of services in the overall ranking table. Default: 20.

categories (string[]): Focus categories for deep-dive sections. Omit for the default set.

Example

analyze({
  "mode": "aeo_article",
  "quarter": "Q2 2026",
  "format": "markdown",
  "article_top_n": 30,
  "categories": ["accounting", "hr", "crm"]
})

analyze({
  "article_type": "quarterly_ranking",
  "target_keyword": "Japanese SaaS agent readiness"
})

Mode Auto-Detection

Priority	Condition	Mode
1	mode parameter is set	(explicit override)
2	category alone (no aeo_service_id, no article_type/target_keyword)	token_savings
3	cost_service_id is present	cost
4	period + model both present	cost
5	aeo_service_id (without article_type/target_keyword)	aeo_report
6	article_type or target_keyword is present	aeo_article

If no actionable parameters are provided, the tool returns an error listing available modes.

All Parameters

Parameter	Type	Required	Modes	Description
mode	enum	--	all	Explicit mode: token_savings, cost, aeo_report, aeo_article
services	string[]	--	token_savings	Service IDs to analyze
task	string	--	token_savings	Task context for tailored analysis
cost_service_id	string	--	cost	Specific service to audit
period_days	integer	--	cost	Analysis period in days (default: 30)
top_n	integer	--	cost	Max recommendations (default: 10)
min_priority	enum	--	cost	Min priority: low, medium, high
aeo_service_id	string	--	aeo_report	Filter by service ID
category	string	--	token_savings, aeo_report	Filter by category
aeo_top_n	number	--	aeo_report	Top N services (default: 20)
include_recommendations	boolean	--	aeo_report	Include recommendations (default: true)
quarter	string	--	aeo_article	Report period label
format	enum	--	aeo_article	Output: markdown or json
article_top_n	number	--	aeo_article	Ranking table size (default: 20)
categories	string[]	--	aeo_article	Deep-dive categories
article_type	string	--	aeo_article	Article type hint
target_keyword	string	--	aeo_article	Target keyword
model	string	--	cost	Model name hint for cost detection
period	string	--	cost	Period hint for cost detection

Notes

• This is an admin/operator tool primarily used for business intelligence, consulting deliverables, and content marketing.
• Token savings analysis provides concrete ROI numbers for justifying KanseiLink adoption. Use lookup with the returned service_id for the full integration guide.
• Cost audit recommendations are sorted by priority (high first) then by estimated monthly savings. Acting on high-priority recommendations first yields the greatest impact.
• AEO reports can be shared with SaaS vendors to encourage agent-readiness improvements in their APIs.
• AEO articles in markdown format are ready to publish. Review rankings for accuracy before distribution.
• Reporting outcome data (via the report tool with model_name and token counts) improves the accuracy of future cost audits.