← Back to directory
D

Daiso MCP

Community
Connect Korean local retail, convenience info, and cinemas to AI via MCP, CLI, and Codex Skill.
Category
Other #42 of 230
Stars
★ 338 Very popular
Transport
Streamable HTTP
Runtime
Hosted (no local runtime) · Node.js
Credentials
No credential needed
License
MIT
Last commit
Tools
41
64FMRS · C

daiso-mcp is a feature-rich MCP server covering Korean local retail, convenience information, and cinemas, supporting multiple AI apps, but it relies on non-official data and has call limits.

Strongest · Documentation 16/20 Weakest · Reliability 8/20

Reliability
8/20
Security and permissions
12/20
Maintenance
14/20
Documentation
16/20
Setup experience
14/20
Why each score
Reliability 8/20
Only README and repo metadata were supplied; the MCP manifest is not cached, so the tool list and init handshake cannot be verified. README claims CI coverage, health checks and CLI smoke tests, but these are untrusted claims; the server also depends on many external retail/cinema scraping endpoints whose failure modes are not controllable. Happy path is plausible but reproducible evidence is missing, so points are deducted.
Security and permissions 12/20
No real tokens are exposed; keys are documented as env vars or Worker secrets. Data flow is fairly transparent: rate limits, 30-day retention of stats, and authenticated health checks are disclosed. However, source is unavailable to verify least-privilege scoping and explicit confirmation for side-effecting tools like submit_developer_request; a hosted remote server also means third parties observe queries. Deduction is for unverifiable permission boundaries.
Maintenance 14/20
MIT-licensed, not archived, with 316 stars and 9 open issues; README shows recent maintenance signals (analytics dated 2026-07/08, coverage workflow, uptime monitor) and multiple contributors. But no commit history, releases, or security-response channel are available, and dependency updates or issue responsiveness cannot be verified, so a moderately high score is given.
Documentation 16/20
README is very rich: supported services table, MCP/CLI/Codex Skill usage, per-client integration examples, REST/OpenAPI specs, rate-limit and health-check explanations, env var setup and many linked docs. Deductions: the MCP tool parameters and manifest were not supplied with the review material, and advanced capabilities like local deployment and OpenAPI facade depend on in-repo docs that could not be verified.
Setup experience 14/20
The simplest path is clear: add the remote URL https://mcp.aka.page in mainstream clients, or use npx daiso; config examples for Claude, ChatGPT, Home Assistant and Grok are provided, and CLI examples are runnable. Deductions: local or self-hosted deployment is only sketched (Cloudflare Workers + env vars), and no source or CI files are available for verification; per the static calibration, the score is capped at 15.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • AI application users interested in Korean retail and local information
  • Developers who need quick access to convenience store, retail, and cinema inventory and schedules
  • Korean users of ChatGPT, Claude, Home Assistant, or Grok

Not for

  • Enterprise users requiring official API authentication, as this is a community project
  • Users needing information outside of Korea
  • Users who require a real-time, unlimited server (public server has daily call limits)

Required permissions

  • Public server only requires network access, no API keys needed
  • For self-hosted deployments, relevant API keys must be configured (e.g., OPINET_API_KEY, NAVER_CLIENT_ID, etc.)
  • Submitting developer requests requires Supabase credentials

Risks and side effects

  • Public server has daily call limits (3,000 requests per IP per day), which may cause rate limiting
  • Data comes from non-official sources and may be inaccurate or delayed
  • Third-party service API terms may require additional authorization; use at your own risk
  • This server is an unofficial community project and not officially supported

Setup

Before you start

Runtime:Hosted (no local runtime) · Node.js

OPINET_API_KEY optionalsecret Opinet (Korea Petroleum Corporation) fuel price API key, set via env var or Cloudflare Worker Secret when self-hosting; not needed for the hosted endpoint.
GOOGLE_MAPS_API_KEY optionalsecret Needed only when self-hosting, for geocoding the location keyword in Opinet radius searches.
SUPABASE_SERVICE_ROLE_KEY optionalsecret Supabase service role key for the developer request feature; a secret, needed only when self-hosting.
NAVER_CLIENT_SECRET optionalsecret Naver local search API secret; a secret, needed only when self-hosting.
HEALTH_CHECK_SECRET optionalsecret Key required to call GET /api/health/checks and /api/rate-limit/stats endpoints.
Other optional settings (3)
SUPABASE_URL optional Supabase URL for storing developer requests via submit_developer_request; needed only for local runs or deployments.
NAVER_CLIENT_ID optional Naver local search API app ID for place/restaurant/cafe searches; needed only for local runs or deployment.
HEALTH_CHECK_BASE_URL optional Optional base URL used for internal health check requests.

Add the MCP server URL https://mcp.aka.page in your AI application that supports remote MCP. For example, in Claude Code: claude mcp add daiso-mcp https://mcp.aka.page --transport http. Alternatively, use the CLI: npx daiso.

claude_desktop_config.json
{
  "mcpServers": {
    "daiso-mcp": {
      "url": "https://mcp.aka.page"
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "daiso-mcp": {
      "url": "https://mcp.aka.page"
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add --transport http daiso-mcp https://mcp.aka.page

Run it in a terminal; replace any <…> placeholders with your own values first.

Check that it works

Add the remote MCP server URL https://mcp.aka.page to your AI app (Claude, ChatGPT, Home Assistant, etc.), then check that tools like daiso_search_products, places_search_nearby, and cgv_get_timetable appear in the tool list; you can also run `npx daiso health` to confirm the service is up.

Troubleshooting

  1. Check if you have exceeded the daily call limit; wait until the next day or self-host
  2. Visit https://aka-page.betteruptime.com/ to see service status
  3. Use npx daiso health to check CLI health
  4. Ensure network access to https://mcp.aka.page
  5. For self-hosted deployments, verify environment variables are set correctly

Things to try

Once connected, you can ask your AI assistant things like:

  • Search Daiso for storage boxes using daiso_search_products
  • Compare which store sells cola cheapest using daiso_compare_products
  • Find cafes near Gangnam Station using places_search_nearby
  • Show me today's CGV movies and timetable for Gangnam using cgv_get_movies and cgv_get_timetable

Tools 41

daiso_search_products read-only
Search Daiso products.
daiso_get_product read-only
Get Daiso product details.
daiso_search_stores read-only
Search Daiso stores.
daiso_get_inventory read-only
Check Daiso store inventory.
daiso_get_display_location read-only
Get display location of a Daiso product.
daiso_compare_products read-only
Compare product prices across retailers.
places_search_nearby read-only
Search nearby places (restaurants, cafes, etc.).
opinet_get_average_prices read-only
Get national average fuel prices.
Show 33 more tools
opinet_get_lowest_price_stations read-only
Get lowest price gas stations.
opinet_search_stations_around read-only
Search nearby gas stations.
opinet_get_station_detail read-only
Get gas station details.
oliveyoung_search_stores read-only
Search Olive Young stores.
oliveyoung_search_products read-only
Search Olive Young products.
oliveyoung_get_inventory read-only
Check Olive Young store inventory.
gs25_search_stores read-only
Search GS25 stores.
gs25_search_products read-only
Search GS25 products.
gs25_get_inventory read-only
Check GS25 store inventory.
seveneleven_search_stores read-only
Search 7-Eleven stores.
seveneleven_search_products read-only
Search 7-Eleven products.
seveneleven_get_inventory read-only
Check 7-Eleven store inventory.
seveneleven_get_popular_keywords read-only
Get 7-Eleven popular search terms.
seveneleven_get_catalog read-only
Get 7-Eleven catalog.
cu_search_stores read-only
Search CU stores.
cu_get_inventory read-only
Check CU store inventory.
emart24_search_stores read-only
Search Emart24 stores.
emart24_search_products read-only
Search Emart24 products.
emart24_get_inventory read-only
Check Emart24 store inventory.
lottemart_search_stores read-only
Search Lotte Mart stores.
lottemart_search_products read-only
Search Lotte Mart products.
lottemart_get_inventory read-only
Check Lotte Mart store inventory.
cgv_get_theaters read-only
Get CGV theater list.
cgv_get_movies read-only
Get CGV movie list.
cgv_get_timetable read-only
Get CGV movie schedule.
megabox_get_branches read-only
Get Megabox branch list.
megabox_get_movies read-only
Get Megabox movie list.
megabox_get_timetable read-only
Get Megabox movie schedule.
megabox_get_seats read-only
Get Megabox remaining seats.
lottecinema_get_branches read-only
Get Lotte Cinema branch list.
lottecinema_get_movies read-only
Get Lotte Cinema movie list.
lottecinema_get_seats read-only
Get Lotte Cinema remaining seats.
submit_developer_request writes
Submit a feature request or bug report to the developer.

Use cases

Search Daiso products and check inventory
Compare prices of a product across multiple retailers
Find nearby restaurants or cafes
Find nearby gas stations and fuel prices
Check stores and inventory for GS25, 7-Eleven, CU, Emart24
Get movie schedules for CGV, Megabox, Lotte Cinema

Supported clients

Claude DesktopPartial support
Claude Code
ChatGPTPartial support
Home Assistant
GrokPartial support

Listed from the project's documentation, not tested by this site.

Overview

daiso-mcp is an MCP server that provides product, store, inventory, price comparison, nearby places (restaurants, cafes), gas station/oil price lookup, and cinema (CGV, Megabox, Lotte Cinema) schedules for Korean local retail (Daiso, Olive Young, Lotte Mart) and convenience stores (GS25, 7-Eleven, CU, Emart24). Users can use it without preparing API keys. The project also offers a CLI (npx daiso) and a Codex Skill, and supports remote connection over HTTP.

Similar servers

BioMCP 75 · B

One binary. One grammar. Evidence from the biomedical sources you already trust.

★ 654 · Tools 11 Compare with this →

Source revision 89fe0c16ed15 Data synced 2026-10-12 Read the FMRS scoring method