# AXO: agent guide

Diagnoses whether AI agents can orchestrate bundled service pages in banking, mobility, travel and health (choose, compute, judge cancellation, hand off), with a nine-cell offer card, orchestration readiness and guardrail-checked content blocks. Results build a demand dictionary, an offer ledger and action contracts. Markets: Korea and Japan.

## Connect
- MCP: `https://daas-axo.listeningmind.ai/mcp` (Streamable HTTP). Start with `app_guide`.
- Browser agents: tools are registered on every page through WebMCP (`document.modelContext`).
- REST: `https://daas-axo.listeningmind.ai/.well-known/openapi.json`

## Approval
- read and action tools run at once.
- run and delete tools need a person's approval. Over MCP the tool returns `confirm_url`; show it to the person, then call `confirmation_status` with the token until state is done.
- In the browser the page shows the person a confirm box.

## Workflow
1. `run_diagnose` (run): Check one financial product page (100 Lima).
2. `get_job` (read): Read until the status is done, then use the result (type, fact ledger, action contract, scores, section verdicts).
3. `scan_sitemap` (read): For many pages, read a sitemap or an address list first (free).
4. `run_batch` (run): Check the picked pages at once (50 Lima a page).
5. `get_batch` (read): Read each page's status and scores in the batch.
6. `run_site` (run): Check agent access across the whole site (200 Lima).

## Inputs
One financial product page address (page check), many addresses or a sitemap (batch, up to 300), or a site address (whole-site agent access check). Markets kr, jp.

## Outputs
Per page: page type, financial checks, the product fact ledger (rates, fees and terms with as-of date and source), the agent action contract, five scores, keep/improve/add per section with draft fixes, and a preview of the finished page. Batches export CSV.

## Cost and time
Page check 100 Lima, batch 50 Lima a page, whole-site check 200 Lima (2026-11-06). Reading results, lists and sitemaps is free.

## Limits
Open to every member since 2026-11-06. Markets: Korea and Japan. Results are visible to their owner; a result marked public opens for anyone with the link. Batches take up to 300 pages.

## Using the screens (browser agents)
- `describe_screen`: menus, tabs, fields and buttons of the current screen, and what the page is for
- `go_to`: open a menu or page of this app by its label or path
- `fill_field`: type a value into a field by its label (never presses run; runs go through the run tools, which ask the person)
- `read_this_page`: the text and ids of what the person is looking at

## The ListeningMind apps (same account, same tools, call any of them)
Each app has `/mcp`, `/llms.txt`, `/agents.md` and WebMCP on its pages. `family_apps` and `recommend_app` return this list with addresses.
- [ACO](https://daas-aco.listeningmind.ai/llms.txt): Diagnoses whether a product detail page can be cited by AI answers and used by shopping agents (MCP `https://daas-aco.listeningmind.ai/mcp`)
- [AOO](https://daas-aoo.listeningmind.ai/llms.txt): Measures whether AI agents can get real tasks done on a website (MCP `https://daas-aoo.listeningmind.ai/mcp`)
- [Brand DNA](https://brand-dna.listeningmind.ai/llms.txt): Brand DNA (MCP `https://brand-dna.listeningmind.ai/mcp`)
- [Brand Finder](https://daas-brand.listeningmind.ai/llms.txt): Which brands lead search by generation or field, and what their brand assets are (MCP `https://daas-brand.listeningmind.ai/mcp`)
- [Brand Naming Agent](https://daas-naming.listeningmind.ai/llms.txt): Creates and checks brand name candidates grounded in search data (MCP `https://daas-naming.listeningmind.ai/mcp`)
- [Brand Wiki](https://daas-wiki.listeningmind.ai/llms.txt): Turns a brand website into the entities, wiki pages and evidence AI assistants read (MCP `https://daas-wiki.listeningmind.ai/mcp`)
- [CEP Finder](https://daas-cep.listeningmind.ai/llms.txt): In which situations do people start looking for this category (category entry point map) (MCP `https://daas-cep.listeningmind.ai/mcp`)
- [Creative Director](https://daas-creative.listeningmind.ai/llms.txt): Finds purchase situations (CEP) in search data and drafts a creative, an image and a video for each (MCP `https://daas-creative.listeningmind.ai/mcp`)
- [Demand Forecast](https://daas-forecast.listeningmind.ai/llms.txt): What category demand looks like in the next 3-6 months and when to launch or campaign (MCP `https://daas-forecast.listeningmind.ai/mcp`)
- [Funnel Leak Finder](https://daas-funnel.listeningmind.ai/llms.txt): Where category search journeys leak to competitors (MCP `https://daas-funnel.listeningmind.ai/mcp`)
- [GEO Audit](https://daas-geo.listeningmind.ai/llms.txt): Checks whether a web page is ready to be read and cited by search engines and AI answer engines (MCP `https://daas-geo.listeningmind.ai/mcp`)
- [Intent KG](https://daas-intent.listeningmind.ai/llms.txt): What people look for about a topic, why, and how they choose (Intent Knowledge Graph) (MCP `https://daas-intent.listeningmind.ai/mcp`)
- [Intent Panel Survey](https://daas-icss.listeningmind.ai/llms.txt): Survey and interview consumer intent groups that share the same job, as often as you need (MCP `https://daas-icss.listeningmind.ai/mcp`)
- [Next Item](https://daas-pulse.listeningmind.ai/llms.txt): What is rising in search for a gender and age segment right now (MCP `https://daas-pulse.listeningmind.ai/mcp`)
- [Prompt Builder](https://daas-prompt.listeningmind.ai/llms.txt): Finds what consumers ask AI and builds managed brand prompts with AI visibility (MCP `https://daas-prompt.listeningmind.ai/mcp`)
- [Smart Store Suite](https://daas-store.listeningmind.ai/llms.txt): Which Smart Store products to grow, and how to rewrite their titles and detail pages in searchers' words (MCP `https://daas-store.listeningmind.ai/mcp`)
- [Taste Finder](https://daas-taste.listeningmind.ai/llms.txt): In what sensory and emotional words do people choose this (consumer taste report from search data) (MCP `https://daas-taste.listeningmind.ai/mcp`)
- [ListeningMind Data API](https://daas.listeningmind.ai/llms.txt): the search data behind every app (MCP `https://daas.listeningmind.ai/mcp`) · [glossary](https://daas.listeningmind.ai/en/glossary) · [all apps](https://daas.listeningmind.ai/en/apps)

## Run stages
- `fetch`: Fetching the page
- `render`: Opening in a browser
- `site`: Checking the site
- `analyze`: Page checks
- `offer`: Genre, type and offer card
- `intent`: Intent dictionary
- `snapshot`: Page capture
- `orchestrate`: Orchestration readiness and blocks
- `ledger`: Writing the fact ledger

## All tools
### run_diagnose
Diagnoses one service page for agent orchestration across four genres (banking, mobility, travel, health): judges genre and page type with evidence, builds the nine-cell offer card, checks pages against UCP-style fields, scores orchestration readiness over real search scenarios, and drafts guardrail-checked content blocks (also the fact ledger and agent action contract). 100 Lima a run. Returns {job}; read it with get_job (result.axo). Spends ListeningMind Data API credits or paid model calls. Over MCP a person approves it first (you get a confirm link); in the browser the page asks the person. Cost: 100 Lima a run. Time: 1-3 minutes. Usually followed by: get_job.
`POST /api/diagnose`
- `url` (required): page address
- `market`: market: kr (Korea) or jp (Japan)
- `kind`: older page type (product or compare); page_type is preferred and overrides it
- `lang`: language of the report: ko, en or ja
- `genre`: service genre: banking (ABO), mobility (AMO), travel (ATO), health (AHO); empty = judged automatically from the page
- `page_type`: page type: detail, list, bundle (package), rates (rate table), compare, info (guide), support (terms), location, brand; empty = judged automatically

### start_batch_scan
Quick scan of up to 300 service pages: genre, page type, offer card, action contract and rule checks from the server HTML (no models) plus one search-volume lookup per page, then ranks them by opportunity = demand share x (100 - quick score). Free. Pick pages for run_batch from get_batch. Spends ListeningMind Data API credits or paid model calls. Over MCP a person approves it first (you get a confirm link); in the browser the page asks the person.
`POST /api/batch`
- `urls` (required): service page addresses (up to 300)
- `market`: market
- `title`: batch title (usually the site host)
- `genre`: banking, mobility, travel or health; empty to detect per page

### run_batch
Full diagnosis of the pages you pick from a quick-scanned batch (start_batch_scan first). Each page gets the demand dictionary, orchestration readiness, guardrail-checked fixes and the finished page. Charged per page (50 Lima a page, counted from urls); the server runs 2 at a time. Spends ListeningMind Data API credits or paid model calls. Over MCP a person approves it first (you get a confirm link); in the browser the page asks the person. Cost: 50 Lima a page. Usually followed by: get_batch.
`POST /api/batch/{bid}/run`
- `bid` (required): batch id from start_batch_scan
- `urls` (required): page addresses from the batch to diagnose fully
- `ocr`: also read text inside images (slower)

### run_site
Checks how well AI agents can reach a whole financial site (robots, sitemaps, key pages, structured data). 200 Lima a run. Returns {job}; read it with get_job. Spends ListeningMind Data API credits or paid model calls. Over MCP a person approves it first (you get a confirm link); in the browser the page asks the person. Cost: 200 Lima a run. Usually followed by: get_job.
`POST /api/site`
- `url` (required): site address
- `market`: market: kr (Korea) or jp (Japan)
- `lang`: language of the report: ko, en or ja

### scan_sitemap
Reads a sitemap address or pasted addresses and groups the pages, before a batch. Free. Read only, free.
`POST /api/sitemap`
- `raw` (required): sitemap address or addresses, one a line

### pick_sitemap
Picks pages from a read sitemap by group, kind and count. Free. Read only, free.
`POST /api/sitemap/pick`
- `sid` (required): sitemap id returned by scan_sitemap
- `gids`: group numbers to take pages from
- `kinds`: page kinds to keep (product, compare, ...)
- `n`: how many pages to pick
- `mode`: how to pick: spread across groups or take the top ones
- `q`: a word the page address must contain

### get_job
One page or site check with its status and result. Its owner (or a public link) only. Free. Read only, free.
`GET /api/job/{jid}`
- `jid` (required): check id (job) returned by run_diagnose or run_site

### get_job_snap
Screenshot of the diagnosed page with section boxes (JPEG). Free. Read only, free.
`GET /api/job/{jid}/snap.jpg`
- `jid` (required): job id from run_diagnose or run_batch

### get_preview
Preview of the page with the chosen fixes applied (HTML; s = comma-separated fix ids or all). Free. Read only, free.
`GET /p/{jid}`
- `jid` (required): job id from run_diagnose or run_batch

### get_preview_frame
The finished page itself, scripts removed (HTML, shown inside the preview). Free. Read only, free.
`GET /p/{jid}/frame`
- `jid` (required): job id from run_diagnose or run_batch

### list_jobs
The signed-in member's page and site checks, newest first. Free. Read only, free.
`GET /api/jobs`
- `limit`: how many to list (at most 100)

### get_batch
One batch with every page's status and scores. Free. Read only, free.
`GET /api/batch/{bid}`
- `bid` (required): batch id returned by run_batch

### get_batch_csv
Downloads a batch board as CSV (one row a page: status, genre, page type, scores). Free. Read only, free.
`GET /api/batch/{bid}/csv`
- `bid` (required): batch id from run_batch

### list_batches
The signed-in member's batches. Free. Read only, free.
`GET /api/batches`

### get_regs
Legal guardrail rules AXO applies per genre and market (required notices, prohibited wording, with the law or guideline and its official source), plus the app's own scheduled checks: daily source-change detection, rule self-test and weekly re-check of recently diagnosed pages. Free. Read only, free.
`GET /api/regs`
- `genre`: service genre; empty = all
- `market`: market: kr or jp; empty = both

