Agent skill

Pp Michi No Eki

by mvanhorn in mvanhorn/printing-press-library

Find and compare Japan roadside-station service stops with source evidence and explicit unknowns.

Apache-2.0Auto-check: notes

Install Pp Michi No Eki

skills CLI
$ npx skills add mvanhorn/printing-press-library --skill pp-michi-no-eki -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install mvanhorn/printing-press-library pp-michi-no-eki --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/mvanhorn/printing-press-library.git skills-src && mkdir -p .claude/skills && cp -r skills-src/cli-skills/pp-michi-no-eki .claude/skills/pp-michi-no-eki && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
pp-michi-no-eki
GitHub stars
2.1k
Token cost
~3.5k tokens
SKILL.md length
1,596 words
Files
1
Skills in repo
506
Repo updated
First seen
Licence
Apache-2.0

At a glance

Find and compare Japan roadside-station service stops with source evidence and explicit unknowns.

  • Works in 3 steps: Install via the Printing Press… → Verify: michi-no-eki-pp-cli --version → Ensure the reported install directory is…
  • Phrases: find Nagano roadside stations with onsen
  • SKILL.md covers Prerequisites: Install the CLI, When to Use This CLI, Anti-triggers and Unique Capabilities, plus 5 more sections
  • Calls npx and go

What it does

Pp Michi No Eki is an agent skill from mvanhorn/printing-press-library. Find and compare Japan roadside-station service stops with source evidence and explicit unknowns. Trigger phrases: find Nagano roadside stations with onsen, compare roadside-station parking, nearby Michi-no-Eki service stops, check station-linked notices, use michi-no-eki, run michi-no-eki.

Its SKILL.md is about 3.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Official library of CLIs generated by the CLI Printing Press. Endorsed, tested, and community-contributed. The licence is Apache-2.0.

When your agent uses it

  • Phrases: find Nagano roadside stations with onsen
  • Compare roadside-station parking
  • Nearby Michi-no-Eki service stops
  • Check station-linked notices

Example prompts

  • “/pp-michi-no-eki”

Requirements

  • Node.js
  • Pre-approved tools (allowed-tools): Read, Bash

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. Install via the Printing Press installer. It defaults binaries to $HOME/.local/bin on macOS/Linux and…
  2. Verify: michi-no-eki-pp-cli --version
  3. Ensure the reported install directory is on $PATH for the agent/runtime that will invoke this skill.

What it can do on your machine

Read from SKILL.md and the folder at commit d9a1696. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Read
    • Bash

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • npx
    • go

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use npx, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Pp Michi No Eki loads about 3.5k tokens when it runs. Until then it costs about 80 tokens; SKILL.md has 1,596 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~80
When it runs · the whole SKILL.md, loaded when a task matches
~3.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
    allowed-tools: Read, Bash

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from mvanhorn/printing-press-library at commit d9a1696, republished under its Apache-2.0 licence (© mvanhorn). 1,596 words, ~3,534 tokens.

Download SKILL.mdSave it as .claude/skills/pp-michi-no-eki/SKILL.md (or your agent's skills folder).
name
pp-michi-no-eki
description
Find and compare Japan roadside-station service stops with source evidence and explicit unknowns. Trigger phrases: `find Nagano roadside stations with onsen`, `compare roadside-station parking`, `nearby Michi-no-Eki service stops`, `check station-linked notices`, `use michi-no-eki`, `run michi-no-eki`.
allowed-tools
Read, Bash
author
zjsng
license
Apache-2.0
argument-hint
<command> [args]
<!-- GENERATED FILE — DO NOT EDIT.
     This file is a verbatim mirror of library/travel/michi-no-eki/SKILL.md,
     regenerated post-merge by tools/generate-skills/. Hand-edits here are
     silently overwritten on the next regen. Edit the library/ source instead.
     See the repository agent guide, section "Generated artifacts: registry.json, cli-skills/". -->

Michi-no-Eki source workflow

Prerequisites: Install the CLI

This skill drives the michi-no-eki-pp-cli binary. You must verify the CLI is installed before invoking any command from this skill. If it is missing, install it first:

  1. Install via the Printing Press installer. It defaults binaries to $HOME/.local/bin on macOS/Linux and %LOCALAPPDATA%\Programs\PrintingPress\bin on Windows:
    bash
    npx -y @mvanhorn/printing-press-library install michi-no-eki --cli-only
  2. Verify: michi-no-eki-pp-cli --version
  3. Ensure the reported install directory is on $PATH for the agent/runtime that will invoke this skill.

If the npx install fails (no Node, offline, etc.), fall back to a direct Go install (requires Go 1.26.6 or newer). This installs into $GOPATH/bin (default $HOME/go/bin), so add that directory to $PATH instead:

bash
go install github.com/mvanhorn/printing-press-library/library/travel/michi-no-eki/cmd/michi-no-eki-pp-cli@latest

If --version reports "command not found" after install, the runtime cannot see the binary directory on $PATH. Do not proceed with skill commands until verification succeeds.

Search the official Japanese directory by prefecture, source region, and facilities. Compare parking and published hours, rank nearby candidates, inspect bounded station notices, and compare saved observations without inferring camping permission.

When to Use This CLI

Find Japan roadside-station service stops; inspect source-listed food, shops, baths, parking and facilities; compare a shortlist; rank scoped nearby candidates; inspect bounded station-linked notices; compare saved factual observations offline. Resolve filter vocabulary with catalog, then use concrete IDs from find.

Anti-triggers

Use the relevant booking/provider interface for campsite/RV inventory or transactions. Use routing tools for driving/vehicle-clearance planning. Station listing or active icons do not establish overnight lodging, outdoor camping or live service availability.

Unique Capabilities

Native station planning preserves read-only source behavior. Optional local learning has explicit write hints, validates cached resource identity, and preserves supported derived rules on targeted undo.

These commands combine source evidence into bounded planning workflows.

Accountable campervan stops
  • readiness — Inspect source facts and permission or service unknowns for one station ID

    Campervan planner; trip agent can inspect source evidence without merging tabs manually.

    bash
    michi-no-eki-pp-cli readiness 19187 --agent
  • compare — Compare up to six ordered station IDs with source facts and per-ID failures

    Campervan planner; trip agent can inspect source evidence without merging tabs manually.

    bash
    michi-no-eki-pp-cli compare --ids 19187,19189 --agent
  • nearby — Rank source stations by straight-line kilometers from required latitude/longitude, scoped by area and facilities

    Campervan planner can inspect source evidence without merging tabs manually.

    bash
    michi-no-eki-pp-cli nearby --prefecture nagano --facility onsen --lat 35.8632686 --lon 138.2769108 --limit 3 --agent
  • station-notices — Find notices explicitly linked to one station ID within page/detail scan limits

    Advisory researcher; trip agent can inspect source evidence without merging tabs manually.

    bash
    michi-no-eki-pp-cli station-notices 19487 --max-scan-pages 1 --max-detail-records 10 --agent
  • changes — Compare two saved snapshot files offline by stable station ID and factual fields, or use an explicit synthetic demo

    Trip agent; advisory researcher can inspect source evidence without merging tabs manually.

    bash
    michi-no-eki-pp-cli changes --demo --agent

Source workflow

  1. Run catalog for official IDs/aliases. Choose --prefecture or the provider's --region, not both. Preserve Japanese names and canonical URLs.
  2. Use find --facility with explicit --match all|any. Provider multi-facility search is any-of; the default all-of is a local required-facility filter. present is listed evidence, not_listed is an inactive source icon, and missing data is unknown.
  3. Use station, compare or readiness for published hours, nominal parking capacities and operator handoff links. Current opening, fees, charging operation, spaces and vehicle fit remain unknown. General MLIT rest guidance stays separate from per-station permissions.
  4. Use nearby only with supplied itinerary coordinates; distances are straight-line kilometers. Inspect candidate cap/coverage and missing-coordinate warnings.
  5. Use notices or station-notices with independent page/detail scan caps. Match exact linked station IDs. station-notices retains every notice on the scanned index pages, then opens only the --max-detail-records prefix; a no-match covers those opened details, not later index rows. Publication date is a JST calendar date, distinct from dates discussed in text. Follow the canonical notice for full text/attachments.
  6. Use complete snapshot --ids ... --json stdout for user-directed local saving. changes reads two snapshots, excludes observation timestamps, and marks fetch gaps unresolved. Projected outputs cannot serve as complete snapshots.

Command Reference

Use find, station, notices, notice, catalog, guidance, snapshot, and the verified capability commands above. Each command exposes domain examples in help. Low-level stations search/get and bulletins list/get return HTML metadata/links; prefer domain commands for planning.

Recipes

Inspect optional local learning evidence
bash
michi-no-eki-pp-cli recall "station 19187" --agent --debug-mismatches
michi-no-eki-pp-cli learnings list --agent

# Only when the user asks to undo this specific local teaching
michi-no-eki-pp-cli learnings forget "station 19187" --resource 19187 --agent

These optional helpers write local learning/audit state; they do not fetch station evidence or populate a nationwide station cache. A known conflicting cached resource remains in mismatches and cannot become an exact hit through a teaching alias or synthesized pattern. Direct aliases require a canonical shared by query, teaching and cached resource. Pattern validation uses the exact entity that produced its candidate, not unrelated entities elsewhere in the query. Recall considers later verified entity bindings from the same pattern when an earlier binding conflicts, retains its first accepted binding, and accepted typed resource IDs are omitted from mismatch diagnostics and their warnings. Final typed-ID deduplication retains the best validated hit under Recall ranking, so a prior partial or lower-ranked hit cannot hide better evidence. recall --limit caps final results, not local database work: candidate validation can scan the full optional pattern store to preserve identity filtering and final ranking. Large local stores can take longer; no local-work or latency bound is claimed. Identifier-verified patterns with no extractable cached identity retain their legacy matching behavior; that is not evidence of current provider identity or facts. Missing cached resources retain an explicit warning and still require a live source fetch. Only a genuine missing row uses this fallback; unexpected cached-identity read failures or cancellation fail recall with runtime exit 5 and no success envelope. Undo reconciles only affected inferred families. A rule needs two distinct compatible positive (boost) examples from eligible teaching sources; unusable rows do not veto valid support. New synthesis and reconciliation after an affected forget exclude hide/alias effects. Unsupported affected rules are removed; explicitly taught and unrelated patterns remain. An explicit teach-pattern records its full declared scope; later inference cannot replace it.

Native domain calls and low-level stations/bulletins reads suppress automatic learning journals/corrections. The six stateful MCP helpers disclose local writes; normal optional recall results remain available.

Show full SKILL.md (628 more words)Show less
Compare stop evidence
bash
michi-no-eki-pp-cli compare --ids 19187,19189 --agent --select stations.id,stations.name,stations.parking,fetch_failures

Return ordered station facts and explicit fetch gaps.

Nearby bath stops
bash
michi-no-eki-pp-cli nearby --prefecture nagano --facility onsen --lat 35.8632686 --lon 138.2769108 --limit 3 --agent

Straight-line distance ranks scoped source candidates, not driving routes.

Station notice watch
bash
michi-no-eki-pp-cli station-notices 19487 --max-scan-pages 1 --max-detail-records 10 --agent

Match exact linked station IDs within a bounded notice scan.

Offline observation changes
bash
michi-no-eki-pp-cli changes --before before.json --after after.json --agent

Compare user-saved factual observations without network.

Auth Setup

Authorized directory routes require no authentication. No accounts, cookies or bookings are used.

Run michi-no-eki-pp-cli doctor to verify setup.

Output and bounds

Use --agent/--json; project with --select to reduce response size. Empty arrays are []. find/nearby return at most50 stations and separately cap candidate examination at5000 (default2000). Compare/snapshot accept at most6 IDs. Notice index scans cap at5 pages and retain every notice on those pages; station-notice detail fetches then open at most50 of them (default10). Rows past that detail prefix are not matches. Declared HTML transport bounds wire bytes and each decode layer to 5 MiB; the domain parser rechecks that bound before parsing. This bounds body bytes, not total process RSS. Snapshot files are capped at 2 MiB. Root --timeout bounds the live workflow, generated pacing applies, and domain observations bypass response cache. Live domain commands reject local mode; changes rejects live mode.

Partial failures appear in fetch_failures and stderr; all-failure operations exit nonzero. The optional MCP mirrors for compare, snapshot and station-notices preserve bounded structured failure accounting in an explicitly failed tool result; they do not rerun the source operation. General exit codes:0 success,2 usage/input,3 not found,4 auth,5 API/runtime,7 rate limiting,10 configuration. Never turn a directory icon, headline count or empty notice scan into permission/inventory/status certainty.

Use the domain CLI workflows for source extraction. Typed MCP HTML tools are hidden by the source spec because their generated handlers expose raw HTML; use the domain CLI or safe command mirrors. This public provider needs no credentials. Credential-like headers such as Authorization and X-API-Key are withheld on a cross-origin redirect; do not add other sensitive custom headers.

Generated low-level link rows can place an HTML page URL in image when no image is published; treat that field as unverified. Use normalized domain workflows for service evidence. Detail endpoint commands return page metadata rather than an empty link-only result.

Low-level detail metadata includes the requested numeric ID, canonical handoff URL, source URL and JST observation time. entity_fields_status distinguishes recognized native identity fields from generic page metadata with explicit unknowns. Domain station/notice remain the richer factual workflows.

Optional framework SQL/workflow-status descriptions mention sync, but this source has no domain sync or database-ingestion workflow; use native domain commands and saved snapshots. Explicit optional learning helpers can open/create/migrate local stores or append audit/telemetry. Six MCP helpers (recall, learnings candidates/list/stats, playbook list and workflow status) declare those local writes with readOnly=false; direct SQL/context remain read-only. Native domain commands, low-level stations/bulletins reads and help suppress automatic journals/corrections. --no-learn does not make explicit optional learning helpers write-free. CLI-only installation does not activate the optional MCP server.

export bulletins writes parsed notice records from the same notice parser as notices and notice, as JSON or JSONL. --limit 0 follows the notice index until it ends; a positive --limit stops once that many records are collected. If another index page remains after 500 pages, export fails instead of writing a partial file. A failed export leaves an existing --output file unchanged. It does not emit the raw HTML page. Use snapshot --ids ... --json for retained factual station observations.

Low-level list/search reads in --data-source auto may initialize or migrate the optional resolver cache even when an HTML response yields no cacheable JSON records. --no-cache controls the HTTP cache; use --data-source live --no-cache to bypass resolver write-through and HTTP caching. The detail metadata adapters use direct source reads. Low-level groups suppress automatic learning journals/corrections; this cache behavior is separate from the twelve native commands and help paths that leave no automatic local state.

© mvanhorn, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in cli-skills/pp-michi-no-eki of mvanhorn/printing-press-library.

Open the folder on GitHubat commit d9a1696

Compare with similar skills

Pp Michi No Eki next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Pp Michi No Eki compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Pp Michi No Eki this skillmvanhorn/printing-press-library2.1k—~3.5kAutomated safety check: NotesApache-2.0
Verify and StopJuliusBrussee/caveman111k1 repos~176Automated safety check: PassApache-2.0
Stop Wordsthedaviddias/Front-End-Checklist74k—~521Automated safety check: PassMIT
Geo Comparesickn33/agentic-awesome-skills47k1 repos~2.7kAutomated safety check: NotesMIT
Compare Python TypecheckersJetBrains/intellij-community21k—~617Automated safety check: PassCustom licence
Recipe Compare Sheet Tabsgoogleworkspace/cli31k—~182Automated safety check: PassApache-2.0

Similar skills

  • Verify and Stop

    JuliusBrussee/caveman

    Prove existing work meets acceptance conditions without expanding scope. Use for validation-only tasks, completion checks, focused gate runs, and last-mile…

    111k GitHub starsUsed in 1 repo~176 tokens
    Agent WorkflowsAuto-check passed
  • Stop Words

    thedaviddias/Front-End-Checklist

    A skill your agent uses when applies primarily when creating new content pages.

    74k GitHub stars~521 tokensUpdated 4 days ago
    Auto-check passed
  • Geo Compare

    sickn33/agentic-awesome-skills

    Monthly delta tracking and progress reporting for GEO clients.

    47k GitHub starsUsed in 1 repo~2.7k tokens
    Auto-check: notes
  • Compare Python Typecheckers

    JetBrains/intellij-community

    Official

    Compare Python type inference with ty, pyrefly, and basedpyright.

    21k GitHub stars~617 tokensUpdated today
    Auto-check passed
  • Recipe Compare Sheet Tabs

    googleworkspace/cli

    Read data from two tabs in a Google Sheet to compare and identify differences.

    31k GitHub stars~182 tokensUpdated 4 days ago
    Documents & OfficeAuto-check passed
  • Compare

    taishi-i/awesome-japanese-nlp-resources

    Compare several Japanese NLP libraries, models, or datasets for a keyword (a specific tool name, or a function/task like '形態素解析') across a handful of criteria chosen for that comparison, rendered as…

    1k GitHub stars~4.1k tokensUpdated 4 days ago
    AI & LLM EngineeringAuto-check: notes

More from mvanhorn/printing-press-library

All 506 skills in this repo
  • Agent Desktop

    mvanhorn/printing-press-library

    Desktop automation through the real Rust agent-desktop CLI, published in Printing Press through a small bridge.

    2.1k GitHub stars~2.3k tokensUpdated today
    Auto-check: notes
  • Gfonts

    mvanhorn/printing-press-library

    Search, browse, and download Google Fonts from the terminal via the gfonts CLI.

    2.1k GitHub stars~574 tokensUpdated today
    Auto-check passed
  • Pp 1688

    mvanhorn/printing-press-library

    The free, offline Trigger phrases: search 1688 for, find a factory on 1688 for, wholesale price on 1688 for, who is the cheapest supplier on 1688 for, compare 1688 suppliers for, use 1688, run 1688.

    2.1k GitHub stars~3k tokensUpdated today
    Auto-check: notes
  • Pp Activity Japan

    mvanhorn/printing-press-library

    Inspect known Activity Japan plan IDs or URLs, compare dated prices and sessions, check language-sitemap coverage, and hand off to canonical booking pages.

    2.1k GitHub stars~2k tokensUpdated today
    Auto-check: notes
  • Pp Adminbyrequest

    mvanhorn/printing-press-library

    Every Admin By Request portal action, plus a local SQLite mirror of audit, events, inventory and requests for ad-hoc...

    2.1k GitHub stars~3.3k tokensUpdated today
    Auto-check: notes
  • Pp Agent Capture

    mvanhorn/printing-press-library

    macOS screen capture, window recording, GIF conversion, and agent evidence bundles from the terminal.

    2.1k GitHub stars~1.6k tokensUpdated today
    Auto-check: notes

Questions about Pp Michi No Eki

What does Pp Michi No Eki do?

Find and compare Japan roadside-station service stops with source evidence and explicit unknowns. Pp Michi No Eki is an agent skill from mvanhorn/printing-press-library. Find and compare Japan roadside-station service stops with source evidence and explicit unknowns.

When should I use Pp Michi No Eki?

Pp Michi No Eki fits situations like: phrases: find Nagano roadside stations with onsen; compare roadside-station parking; nearby Michi-no-Eki service stops; check station-linked notices.

How do I install Pp Michi No Eki in Claude Code?

Run `npx skills add mvanhorn/printing-press-library --skill pp-michi-no-eki -a claude-code`. Or copy the skill folder (cli-skills/pp-michi-no-eki in mvanhorn/printing-press-library) into .claude/skills/pp-michi-no-eki in your project. Claude Code loads it when a task matches its description.

How do I install Pp Michi No Eki in Codex?

Run `npx skills add mvanhorn/printing-press-library --skill pp-michi-no-eki -a codex`. Or copy the skill folder (cli-skills/pp-michi-no-eki in mvanhorn/printing-press-library) into .agents/skills/pp-michi-no-eki in your project. Codex loads it when a task matches its description.

Can I use Pp Michi No Eki in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add mvanhorn/printing-press-library --skill pp-michi-no-eki -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/pp-michi-no-eki, .gemini/skills/pp-michi-no-eki, .github/skills/pp-michi-no-eki and .opencode/skills/pp-michi-no-eki in your project.

What does Pp Michi No Eki need to run?

Going by SKILL.md and its folder, Pp Michi No Eki needs the command-line tools its instructions call (npx and go). Our summary lists: Node.js. Its frontmatter pre-approves these tools: Read, Bash.

Does Pp Michi No Eki access the network?

SKILL.md contains no URLs. Its commands use npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Pp Michi No Eki safe to install?

Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Pp Michi No Eki use?

Pp Michi No Eki is published under the Apache-2.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Pp Michi No Eki use?

About 3.5k tokens (SKILL.md is roughly 14k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Pp Michi No Eki?

Skills that share tags, products or a category with Pp Michi No Eki: Verify and Stop (JuliusBrussee/caveman, 111k stars), Stop Words (thedaviddias/Front-End-Checklist, 74k stars), Geo Compare (sickn33/agentic-awesome-skills, 47k stars) and Compare Python Typecheckers (JetBrains/intellij-community, 21k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Pp Michi No Eki?

mvanhorn (a GitHub user) maintains it in mvanhorn/printing-press-library, which has 2,056 GitHub stars. The repository holds 506 skills in this directory. The repository was last updated on October 9, 2026.

Source: mvanhorn/printing-press-library on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.