Commit graph

53 commits

Author SHA1 Message Date
Wolfgang Nagele
7438fd1065 Agent mode: an AI agent plays in the player's own tab, with the agent's clock or a running one
The MCP server gains a second session, --browser: the game in the player's tab, reached through the
agent relay npm run dev hosts at /agent (loopback unless --host; Host and Origin checked, no token).
The player turns it on in Settings → Agent API, a section only the development page has, and picks the
clock: a pace of 30 s to 5 min a day that nobody can pause, or the agent's own. Headless gets the
running clock too (--clock realtime --pace MS) and both get a wait tool.

All of it lives in src/agent/ (development page only). The game offers HOOKS, UI.clockLocked,
setSpeed() and runClock(dt), the frame's day block moved out unchanged, and nothing more: the page
players get grows by under 1 KB gzipped and runs exactly as before.
2026-11-03 16:18:43 -05:00
Wolfgang Nagele
2526dd458d MCP server on the official SDK, over stdio or HTTP
scripts/mcp.js registers the same tools on @modelcontextprotocol/server (2.4.0, with
@modelcontextprotocol/node 2.1.2 for node:http; dev dependencies, pinned) instead of its own JSON-RPC
loop. stdio works as before; --http [HOST:]PORT serves Streamable HTTP at /mcp, on loopback unless
told otherwise.

Over HTTP every request's Host must be an address of this machine (or --allow-host), an Origin other
than localhost is refused, and a bearer token (SOT_MCP_TOKEN or --token-file) is optional; bound
beyond loopback without one, the server warns. load_save and export_save take JSON only there, or files
under --save-dir and nowhere else. Calls from every client run one at a time, in order.
2026-11-03 16:18:38 -05:00
Wolfgang Nagele
5b76685e5c MCP: the agent facade moves into src/agent/, assembled into the development page only
The command tables and the page-side reads, screens and dispatch (scripts/lib/mcp-commands.js and
mcp-page.js) become src/agent/, the facade window.TUBES.call(tool, args) in the page's own scope.
A new <!-- @include-dev agent --> line in src/index.html pastes it in after the game, but only in the
development page (npm run dev, the MCP server, i18n-check, tests that ask with load({agent:true})).
The build drops the line whole: _site/ is byte for byte what main builds.

Tests load the page players get by default, the build test fails if anything outside src/agent/ names
the facade's api/API_ names, and the pages test checks the built page carries none of it and still
plays. ADR 0004 records the rule.
2026-11-03 16:18:38 -05:00
Frederic Jung
894307a35c
Every capital in the Middle East, the Caucasus and Central Asia, with Bahrain on the Gulf's cables
The MCP server reads a dropdown as its selected option, so a long network list no longer crowds the traceroute out of a report.
2026-11-03 20:48:15 +01:00
Charlie Allom
c5c5aac970
Merge remote-tracking branch 'origin/main' into keyboard-shortcuts 2026-11-03 10:48:01 +00:00
Charlie Allom
3192f42b5a
$ steps through the open offers, soonest to expire first
$ moves the map to the next offer's city, in the Deals list's order,
opens the Customers tab on a POP and marks the offer under Offers here
(UI.offerAt, the existing .row.on style). Two offers in one city each
get a press; any other selection clears the mark. With nothing on the
table it says so with the Deals tab's own "No open offers" text.
2026-11-03 10:47:51 +00:00
Charlie Allom
daec638bfc
MCP: classify popStep, zoomStep and keys, the keyboard shortcuts' actions, as views 2026-11-03 10:39:00 +00:00
Frederic Jung
823e4188be
Pay to hurry the permits and factory of your own subsea cable (closes #258)
Before laying starts, seaRush buys 30 days a step (or all of them at once)
at $500k plus $60 a km, with the market's build prices. The floor is the
factory: 60 days plus a day per ~80 km from the order (prepMin, through
leadDays), then the lay ships still need their time at sea, which a second
ship still halves. 6,000 km: from 600 days of prep to 132 for about $13M.
2026-11-03 10:35:16 +00:00
Frederic Jung
2a503b223e
MCP: fix the quote that broke the rules reference 2026-11-03 10:23:12 +00:00
Frederic Jung
83740911e8
MCP: new_game takes the Best Effort preset (soft); docs and the rules reference say so 2026-11-03 10:22:30 +00:00
Frederic Jung
186e4c65d1
MCP: classify routingOpen, the new Routing & DDoS window, as a view 2026-11-03 10:18:52 +00:00
Frederic Jung
1c498734b4
Merge branch 'pr264' into HEAD 2026-11-03 09:56:14 +00:00
Frederic Jung
3048c18b14
Fit the left tabs on small screens; move routing settings into a modal
The left panel's four tabs overflowed its 272px width on tablets and small
laptops: Goals wrapped its count, Networks ran off the edge. Counts now sit
on the tab's corner, labels stay on one line and shrink last.

Backbone, Auto-TE and DDoS settings sat below every notification in the NOC
tab. A Routing & DDoS button at the top opens them in a modal that redraws
in place after each action, with a red dot when Auto-TE is paused.
2026-11-03 10:20:57 +01:00
Wolfgang Nagele
a72cb7d760 MCP server: typed, checked actions with outcomes and dry-run prices; schema, catalog, state, results and ui tools 2026-11-02 21:10:35 -05:00
Frederic Jung
640b65ce9a
Differential random games: main's page and this checkout's play the same seeded game move for move
scripts/route-fuzz.js plays many random games (every mode: presets, sandbox, scenarios,
the weekly, puzzles, architect runs, or a save) on two pages in one process and compares
all of R, R.fl, R.ef, the PNI lists and S after every move and every day, and routing with
the kept tables against routing without them. tests/route-fuzz.test.js plays four.
2026-11-01 21:44:30 +01:00
Frederic Jung
37def9dd6b
The game as a local MCP server: play, inspect and debug it from an AI client
scripts/mcp.js speaks MCP over stdio and loads the game from src/ in jsdom,
seeded, like the tests. Tools: new_game, load_save, export_save, status,
advance (with per-day timings), network, city, find, traceroute,
peering_check, hot_spots, actions, act, screen, panel, click, open_link.
Actions go through ACT as buttons do; files, slots, settings and the clock
stay out of reach, and nothing evaluates code in the page. Nothing is
hosted and the page is unchanged. See docs/MCP.md.
2026-11-01 18:04:22 +00:00
Frederic Jung
2d1f48f3e6
CI: tests in four parts on four runners, every core in use, older PR runs cancelled
npm test ran every file one after another (--test-concurrency=1, from when there were three files). It now runs as many files at once as there are cores (TEST_JOBS=1 for the old way), and --shard i/n picks one of n parts of about equal time, balanced by tests/durations.json (npm run test:timings). CI runs four parts and the build side by side; a final 'test' job is green only when all are, so the check keeps its name. A newer push to a PR cancels the older run.
2026-10-31 14:11:02 +01:00
Frederic Jung
3e7bbdffad
DDoS rework: tests for the new model, Flood Season and its weekly at level 4, routing reference fix 2026-10-31 12:51:15 +01:00
Frederic Jung
abd55c0846
Release notes: a story for big releases, and the 2.0 story
A CHANGELOG section may open with prose before its first bullet. The build bakes it
into the page and version.json, and "What's new" tells it in full with a link to every
change on GitHub instead of the first sentences of five bullets. The 2.0.0 section gets
its story and every change grouped under headings, Architect runs included.
2026-10-30 16:28:19 +01:00
Frederic Jung
5f478dd318
Weekly Architect runs: sites outgrow their tubes, and a countdown
A grow brief raises a site's traffic past what its tubes carry; the reference makes
room by upsizing the busiest tube, or laying a second one where a long cable cannot
grow. Traffic may now reach 380G. The weekly row says when the next run comes out.
2026-10-30 14:07:45 +01:00
Frederic Jung
4e6e8294d2
Architect runs: a generated run every week
scripts/weekly-architect.js picks an HQ in a region the last weeks did not have and
designs each brief against the network a reference architect has built so far, so
every brief asks for something new; the game's own checklist and chaos monkey review
judge it, and the whole run is replayed from the file for three stars. The game
fetches architect/weekly/<key>.json when Puzzles opens, checks it field by field and
writes the briefs' text itself. The Sunday job and the release deploy carry it like
the weekly puzzles and scenarios. Later briefs' cities no longer show early.
2026-10-30 13:44:04 +01:00
Frederic Jung
92536606dd
Tests, CHANGELOG and docs for the peering desk 2026-10-29 21:41:28 +01:00
Frederic Jung
47c3674e8c
Tests: Gangnam reference hauls from Tokyo; Seoul-only, Tokyo-only and the old plan fail
The reference player learns single PNIs and last-resort transit.
2026-10-29 20:59:17 +01:00
Frederic Jung
0ca9ad8a95
Scenario playtest test: a reference strategy and a failing shortcut for every scenario (C1) 2026-10-29 19:57:15 +01:00
Frederic Jung
d8df1e37f6
Weekly scenario generator with a reference player, and weeks 2026-W44 to W46 2026-10-29 17:34:38 +01:00
Frederic Jung
b640ab9814
Weekly puzzle: schema 2 with traffic, sized references, regenerated weeks
The generator gives every week content in its first city and eyeballs in the
others (a latency sprint adds a circuit with a cap), sizes both references with
the game's routing, and keeps the per-week rev refresh working with it.
2026-10-29 10:50:04 +01:00
Frederic Jung
69689f18f4
Merge v2 (small fixes, weekly rev) into puzzles with traffic 2026-10-29 10:36:21 +01:00
Frederic Jung
e90f5f675e
Puzzles: real traffic routed by the game, routers, link sizes and paths
Every puzzle now carries demand (eyeballs pulling from a content origin, circuits
with latency caps, best-effort flows) routed by compute(). Stars: delivered,
within budget, N-1 under load. Routers, ports, metrics, standby and the
redundancy check work in puzzles; the whole world is open. Weekly schema 2
carries the demand; scripts/lib/puzzle-solver.js sizes and searches builds.
2026-10-29 10:35:28 +01:00
Frederic Jung
7cbe0ec09b
Regenerate published weekly puzzles when a release's prices break them
--check now re-solves every week not over yet and, when its budget no longer
verifies, regenerates it deterministically: the same puzzle with a new budget
if that works, else a fresh one for the key. A regenerated file carries rev 2,
3...; the game keeps stars per revision, so only that week's stars reset.
Past weeks are left alone unless the game rejects them. release.yml runs it
before the tests on every deploy, weekly-puzzle.yml before making the new
week; either fails only if a week cannot be regenerated.
2026-10-29 10:03:20 +01:00
Frederic Jung
8d72259500
Weekly puzzle generator: v2 counts a whole cable as one failure, so require it instead of patching it in 2026-10-29 07:00:34 +01:00
Frederic Jung
372893ec75
Weekly puzzle generator: themed, solver-budgeted, verified by the game's own stars; first three weeks 2026-10-28 20:42:51 +01:00
Frederic Jung
a4f89491e6
Docs: the single-file page and English keys lasted until 1.34 2026-10-27 11:28:17 +01:00
Frederic Jung
0057e74472
Ship short ids instead of labels; checker, tests and docs for labels
The build swaps every label for a short id in the shipped page, the inlined
English and the translation files (scripts/lib/compile-labels.js). Labels
passed to N_() stay whole because saves keep them. Against 1.33: page +4%
gzipped (321 -> 335 KB), translations -40% (it.js 143 -> 85 KB); routing,
day, panels, dashboard, map and boot time unchanged.

- i18n-check rejects labels the build cannot see (outside tr/trn/N_)
- tests/pages.test.js plays the shipped page in English, Italian and French
- docs/adr/0003, add-strings skill, ARCHITECTURE, README, CONTRIBUTING,
  CLAUDE.md describe labels
2026-10-27 11:27:11 +01:00
Frederic Jung
9043af5ab0
port-branch: keep English for labels your side still uses 2026-10-27 11:06:29 +01:00
Frederic Jung
315ab1fac5
port-branch: merge English entry by entry, like the translations 2026-10-27 11:06:11 +01:00
Frederic Jung
76d3d7d199
port-branch: compare plural entries by value 2026-10-27 11:05:35 +01:00
Frederic Jung
e817beed1a
Translations keyed by labels instead of English text
tr('pnis.delivery_2_to_6_weeks') instead of tr('Delivery: 2 to 6 weeks'): the
English lives once, in src/i18n/en.js (inlined into the page at build, so no
extra request), and it/pt/fr map the same labels. Changing English wording no
longer renames a key in every dictionary.

- 2,899 labels, made from where each text is used and its first words; short
  shared words are common.*
- plurals are one label with {one, other}: trn('label', n)
- static markup: data-i18n="label", data-i18n-aria-label, data-i18n-title
- saves from 1.33 and older stored English text in a few places; tr() still
  resolves it
- i18n-check: undefined/unused labels, markup drift, plural shape; i18n-add.js
  works on labels and --rename updates src/ too
- the old-branch port tool gives old English-keyed code and dictionaries the
  same labels
2026-10-27 11:05:09 +01:00
Frederic Jung
6f95539b2d
port-branch: work in both merge directions 2026-10-27 10:23:33 +01:00
Frederic Jung
052fdd713f
port-branch: keep code after the ACT block in very old versions 2026-10-27 10:11:07 +01:00
Frederic Jung
657e7b8083
port-branch: also handles branches started from older versions of the page 2026-10-27 10:10:56 +01:00
Frederic Jung
c64b139a86
Fewer merge conflicts: actions by area, one city per line, sorted dictionaries
- ACT is split by area into game/90-actions/*.js (const ACT in the first file,
  Object.assign in the rest)
- CITY_LL, CITY_NUDGE and LANDS_LL have one entry per line; the id-keyed ones
  are sorted
- site/i18n/*.js are sorted, one entry per line; i18n-check enforces it and
  --fix (npm run i18n:fix) restores it; scripts/i18n-add.js adds, removes
  and renames entries in place
- scripts/legacy/port-branch.js moves a branch's site/index.html edits into
  src/ during a merge with main
- perf workflow and PR template build the base page with scripts/lib/page.js
2026-10-27 10:10:16 +01:00
Frederic Jung
67a40942b4
Split the game page into partials under src/, assembled into one index.html at build
site/index.html (900 KB of HTML, CSS and JS in one file) becomes src/: a shell
(src/index.html), styles/ and game/, concatenated in path order by
scripts/lib/page.js. The assembled page is byte-identical to the old file, so
load time and in-game performance are unchanged.

- scripts/build.js replaces build.sh (same env vars, same output)
- npm run dev: a zero-dependency dev server that assembles on every request
- tests assemble the page in memory; npm test runs every tests/*.test.js
- perf-bench takes --base-ref to build a git ref's page
- scripts/legacy/split-index-html.js is the splitter, kept for older branches
2026-10-27 10:05:56 +01:00
Frederic Jung
c76dd4d556
Benchmark in real Chromium too, and only call a row slower when the samples say so
npm run perf -- --browser times the same workloads in Chromium as a desktop
and as a phone (390x844 at 3x, CPU slowed 4x), one browser per version. Each
run collects garbage first and ends by forcing layout and the map canvas, so
deferred work counts. The jsdom DOM is about 30x slower than a browser's and
has no layout, so the interface rows need this.

Shared machines swing single runs by 30% on identical code, CPU time
included. A row now counts as slower or faster only when it moved 15% and a
Mann-Whitney U test gives p < 0.01 (15 runs a side by default). Identical
code: no flags in three full runs; a routing slowdown planted in a copy is
flagged in jsdom and on the phone.

The panel workload renders both selections every run instead of alternating,
which split its samples in two. CI drives the runner's Chrome through
playwright-core, no browser download.
2026-10-26 12:04:57 +01:00
Frederic Jung
e5deb90968
Routing reference test runs on today's routing loop; perf bench reads exported saves
The reference test from #106 now patches the loop as it looks after #96
(business-first passes) and also covers prefer, prepend and protect
customers. It catches an exit index that ignores RouteScript preferences,
which the shortcut test in speed.test.js cannot see.

npm run perf -- --save accepts files from the in-game Export save, so
scripts/benchmark-performance.js and its README section are not needed.
The performance skill documents --save, both equivalence tests and how to
handle noisy runs.
2026-10-26 08:40:20 +01:00
LookedPath
ffb8c50895 Add routing reference regression and exported-save benchmark 2026-10-26 06:11:12 +00:00
Frederic Jung
308c3f7045
Performance benchmark on a big network, posted on every PR
scripts/perf-bench.js times routing, a game day, an action, panels, the
NOC dashboard and the map on tests/fixtures/big-network.json (104 POPs,
950 PNIs, 832 caches), base and head in alternating turns, and says
whether routing results changed. perf.yml runs it on every PR against
the base; perf-comment.yml posts the table as one updated comment.
New performance skill, CLAUDE.md rule, PR template and docs.
2026-10-26 06:39:23 +01:00
Frederic Jung
4e95b9b397
CI: translation coverage job
A separate 'translations' job runs the i18n checker on every push and
PR and writes a coverage table per language to the run summary, with
the missing, broken and unused keys listed. The checker now also fails
when a language file is not offered in SOT_LANGS or cached in sw.js.
2026-10-26 06:04:28 +01:00
Frederic Jung
5786f76ed9
Rework the phone layout and make the game installable
Phones get a one-row status strip, a bottom tab bar (play, speed, goals,
deals, NOC, networks) and draggable bottom sheets; in landscape the sheets
slide in from the side. Toasts can be tapped away, dialogs become bottom
sheets and the title screen keeps its start button in reach.

Adds a service worker for offline play and an install button (the
browser prompt on Android, instructions on iOS).
2026-10-24 22:52:55 +02:00
Danilo Carbonaro
42711456e2 Tidy the i18n layer after review
- Badges that react to sharks, anchors, crocodiles and llamas read the
  English text of the incident again, so they unlock in every language.
- Scenario names are looked up when shown, so they follow a language switch.
- Fall back to English (and lang="en") when the dictionary did not load,
  and version the dictionary URL so a release never pairs with a stale one.
- Share the segmented-button helper between settings and the title screen,
  merge keys that only differed in a placeholder name, and slim the check
  script.
- Tests: guard the hooks the tour patches into fireFailure/fireEvent, and
  cover the dictionary fallback and the scenario name.
2026-10-24 21:52:27 +02:00
Danilo Carbonaro
04eb32e0a5 Add Italian translation and a language switch
English stays the source text and the default. Player-facing strings go
through tr()/trn(), which look the English up in a per-language dictionary
(site/i18n/<lang>.js) and fall back to it unchanged. The dictionary is only
loaded when the game runs in that language. The language follows the
browser on the first visit and can be switched under Settings or on the
title screen; the choice is remembered.

- site/i18n/it.js: complete Italian translation (~1900 entries)
- scripts/i18n-check.js / npm run i18n:check: lists missing or stale entries
  and translations that drop a {placeholder} or HTML tag
- tests/i18n.test.js: language selection, fallback, dictionary integrity,
  and a tour of every screen, failure and event in English and Italian
2026-10-24 21:52:27 +02:00