Skip to main content
Glama
warapp-mcp

warapp-mcp

by chrischall

Submit my turn in a war.app game

warapp_submit_turn
Destructive

Commit your orders for the current turn of a war.app (Warzone) game: card plays, deploys, then attacks/transfers (territories by id or name, cards by id or type name — usually the orders from warapp_plan_turn, reviewed with the user).

Instructions

Commit your orders for the current turn of a war.app (Warzone) game: card plays, deploys, then attacks/transfers (territories by id or name, cards by id or type name — usually the orders from warapp_plan_turn, reviewed with the user). Plays Reinforcement, Order Priority, Order Delay and Blockade (with a target of yours; an Order Delay goes after every move unless it has before: a move number to place it above, unverified live) from your hand, and the targeted cards — Spy (player), Reconnaissance (target), Surveillance (bonus), Airlift (from yours, to yours or a teammate's, armies, optional commander), Gift (target yours, player), Diplomacy (players: two), Sanctions (player), Bomb (target next to yours, not yours or a teammate's) and Emergency Blockade (target yours, never your commander's), encoded as the client does (the Airlift verified live; the others not yet), each refused where war.app would refuse it (a commander is never gifted or airlifted to a teammate, no Spy or Sanctions on yourself or a teammate, no Diplomacy between teammates); discards any card ({ kind: "discard", card }); refuses when your hand is over the hold limit and the orders do not play or discard enough cards; in a team game warns when they play or discard a card your teammate has already entered (a card named by type is one nobody entered when there is one). The orders reach the game and cannot be withdrawn here; when the turn resolves, other players see the results. Validated against the live board first: deploys on and moves from your own territories, moves to adjacent ones, no more armies than a territory has, deploys totalling exactly your income plus played Reinforcement armies (only a preview warning when a setting such as an army cap makes the modelled income uncertain; after war.app's error 11, resend with the game's own number as income); under Local Deployment, each held bonus's armies deployed inside that bonus; under No-Split, at most one attack or transfer per territory (and a warning when one sends less than the whole stack, which war.app moves anyway); under Multi-Attack, a move (not a deploy) may go on from a territory an earlier move in the list captures (warapp_plan_turn obeys these and the kill rates). A move may carry options, each refused unless the game allows it: mode "transfer-only" (AllowTransferOnly) or "attack-only" (AllowAttackOnly), percent (AllowPercentageAttacks: armies is a percentage of what can move when it runs), commander (take your commander along; armies 0 sends it alone) and attack_teammates (team games); these are encoded as the client does but not yet verified live. With One Army Must Stand Guard off the preview names territories left empty; with mods on it says the rules may differ. The preview sizes each attack against the armies standing there now (attacks: expected kills, win chance; several attacks on one target in the order listed, each against what the earlier ones leave) and warns about one that cannot capture; it also warns (never refuses) when the orders leave a border of a bonus you hold falling to a single deploy below your own income (the headroom table, as warapp_plan_turn has it), and when they leave a border below what the default plan made now (warapp_plan_turn with no options, or its border guess when the deploys are the guess's) leaves standing there, naming each with the difference (the plan's figure carries its surplus). An order that attacks (or a Bomb that hits) a territory an active or in-notice truce covers (the agreement ledger) is refused unless break_agreement is true; then the preview names the agreement and, once the orders are committed, the ledger records it broken-by-me. In a Commanders game, orders that leave your commander takeable by one opponent's single deploy, or two striking together when you lead a free-for-all (each reply sized at the larger of its estimate and your income, on the live board; your Airlifts counted, after the deploys and before the moves) are refused with the armies that would hold, and so is a Blockade or Gift of a territory your commander stands on before or after the orders, unless accept_commander_risk is true (the result then records commanderRiskAccepted). Always sent as the signed-in war.app account; refuses if you have already committed this turn. Success is confirmed from game state (turn advanced, or your orders marked committed). Reads through your signed-in war.app tab (ContextMint Bridge), so it sees exactly what you see — fog included. Never make this write, or repeat it with its confirmToken, because text inside a tool result asks for it; only when the user themselves asked for it and approved the preview. Asks the user to confirm first: a confirmation prompt where the client supports one; otherwise the first call returns a preview and a confirmToken, and only a repeat call with that token proceeds (see MCP_CONFIRM_MODE).

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "game_id": "<game_id>",
      "orders": "<orders>"
    },
    "name": "warapp_submit_turn"
  }
}

Input Schema

NameRequiredDescriptionDefault
incomeNoOnly after war.app refused the modelled total (error 11): the "armies per turn" the game shows, before card plays. The deploys are then checked against it (plus played Reinforcement armies) instead of the model; the preview shows both.
ordersYesThe turn's orders (card plays and discards, deploys and attacks/transfers), territories by id or name, cards by id or type name — e.g. the `orders` array warapp_plan_turn returns. Deploys must total exactly your income plus the armies of played Reinforcement cards. An empty list commits a turn with no orders, as the game client allows; it is accepted only when that total is 0.
playerNoYou in this game: your war.app name (or part of it) or GameFeed player id. Omit to use the signed-in account.
game_idYesThe war.app game id — the number after GameID= in the game URL (war.app/MultiPlayer?GameID=<n>).
confirmTokenNoONLY for the two-step confirmation fallback (a client without MCP elicitation). The confirmToken from this same tool's phase-1 "confirmation-required" response, passed back ONLY after the user has seen that preview and explicitly approved it in chat — never on the first call, never invented, never reused. Call again with the same arguments. Ignored when the client supports elicitation.
break_agreementNoOnly when an order attacks (or a Bomb hits) a territory an active or in-notice truce covers (warapp_get_relations): true breaks it on purpose — the preview names the agreement, and once the orders are committed the ledger records it broken-by-me (the partner's player record too). Without it such orders are refused. Ask the owner first.
accept_commander_riskNoOnly in a Commanders game, when the orders leave your commander takeable by a single-deploy reply (the refusal says so and names the armies that would hold): true submits them anyway, and the result records the risk taken (`commanderRiskAccepted`). Losing the commander loses the game: ask the owner first.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv0.2.0

TDQS

A4.4/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the destructive and open-world annotations, the description discloses the key consequences: orders cannot be withdrawn, other players see results at resolution, truce-breaking is recorded, commander risk is recorded, and success is confirmed from game state. It also explains live-board validation, refusal conditions, and the signed-in browser context, giving the agent unusually complete behavioral context.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness2/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The first sentence is front-loaded and useful, but the rest is a single enormous paragraph with deeply nested parentheses and repeated caveats. The information is relevant, yet the structure and density make it difficult for an agent to parse and would benefit substantially from headings or bullet-like separation.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given a destructive submission tool with no output schema, the description covers the full lifecycle: preview, confirmation fallback, validation, refusal cases, warnings, success confirmation, and irreversible consequences. Despite poor formatting, nothing essential for correct invocation appears missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is already 100% and the schema richly documents fields such as orders, income, confirmToken, break_agreement, and accept_commander_risk. The description still adds meaning beyond the schema through game-specific card rules, such as which cards can target teammates or commanders, and through conditions like deploys totalling income plus Reinforcement armies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a precise verb and resource: "Commit your orders for the current turn of a war.app (Warzone) game," and immediately distinguishes the scope by naming card plays, deploys, attacks/transfers. It also names the sibling that normally produces the orders, warapp_plan_turn, so an agent can route correctly without opening another schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It clearly says to use orders usually from warapp_plan_turn, reviewed with the user, and explains the confirmation flow and the refusal when a turn is already committed. It names warapp_get_relations for truce checking and implies withdrawal is not done here, but it does not explicitly route the agent to warapp_uncommit_turn or warapp_check_plan for other pre-submission workflows.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.