TLDR: Treat a full cedh proxy deck as a controlled copy of one locked decklist. Save the source list, confirm the legal 100-card count, remove maybeboard entries, document any exact-version requests, and keep tokens and experimental swaps in separately labeled groups. When the cards arrive, compare every card with the saved list before sleeving.
Competitive Commander lists change quickly, but your physical deck should not change accidentally. The hard part is rarely finding approximately 100 recognizable cards. It is preserving the exact list you chose while accounting for partner commanders, basic-land quantities, double-faced cards, tokens, alternate treatments, and the handful of speculative swaps hovering around every serious brew.
What counts as a full cEDH proxy deck?
The core deck must contain 100 cards, including its commander. A conventional setup is one commander plus 99 cards in the library. A legal setup with multiple commanders still contains 100 total cards, so the additional command-zone card changes how many cards remain in the library. Commander is also a singleton format apart from basic lands and specific card-level exceptions, and every card must fit the commander’s color identity. Wizards summarizes these construction rules in its official Commander format guidance.
Tokens, emblems, counters, reminder cards, substitute cards for double-faced layouts, and proposed replacements are gameplay accessories rather than permission to exceed the legal deck count. Keep them physically and administratively separate. A package containing 100 deck cards and 12 tokens may have 112 printed pieces, but only the locked 100 constitute the deck.
A complete physical project therefore has two layers: the legal deck and its helper package. Commander Proxies describes its catalog as including ready-to-play Commander decks and custom card builds, but the contents and available customization options can vary. Confirm what a particular ordering method accepts before assuming that tokens, replacement slots, or precise treatments can be bundled with the deck.
Step 1: Freeze one source-of-truth list
Choose one decklist URL or exported file and label it with a date or revision name, such as “Tymna-Kraum 2026-09-10 locked.” That snapshot, not the live page you continue editing at midnight, is the source of truth for the physical deck.
Export a plain-text or CSV copy even if you are submitting a Moxfield or Archidekt link. Deck-hosting pages can change after submission, and an export gives you a stable record for the eventual arrival audit. If you are using Moxfield specifically, the full-deck ordering workflow for Moxfield lists covers the handoff process in greater detail.
- Name and save the locked revision.
- Record every command-zone card separately.
- Confirm that the command zone and library total exactly 100 cards.
- Export the list to a local file without changing its sorting or quantities.
- Save a second list for tokens and optional test cards.
- Do not edit the locked file; create a new revision when the deck changes.
Step 2: Remove everything that is not in the locked 100
Deckbuilding sites encourage experimentation, which is useful until an export quietly includes cards that were never approved for the final list. Check the maybeboard, considering section, side notes, alternate commanders, archived combos, and any custom categories that might appear in the submitted data. Existing complete-deck guidance flags maybeboard leakage, unresolved basic lands, missing token plans, and double-faced-card handling as recurring completion problems.
Then audit quantities. Repeated basic lands may be intentional; two copies of a singleton spell usually indicate an import problem unless that card explicitly permits additional copies. Similar names deserve special attention as well. A list can legitimately contain related cards whose names differ by only a word, so do not “fix” an apparent duplicate without checking the locked source.
The final intake file should contain no placeholders such as “flex counterspell,” “best black tutor,” or “land TBD.” Those are brewing notes, not printable deck entries. Resolve each one to a specific card or leave the order unsubmitted until you do.
Step 3: Build a version manifest
A decklist identifies cards, but it may not communicate which printing, frame, art, language, or layout you want. If presentation matters, attach a separate manifest rather than stuffing ambiguous notes into card names. Ask the provider which identifiers it can honor before relying on collector numbers or set codes.
| Field | Required use | Example format |
|---|---|---|
| Card name | Identifies the game piece | Mana Crypt |
| Quantity | Prevents accidental duplication | 1 |
| Set or treatment | Records a preference when supported | Standard frame; specified set code |
| Collector number | Disambiguates artwork when supported | Set code plus collector number |
| Layout note | Explains a functional presentation need | Double-faced card; both faces required |
| Priority | Separates requirements from preferences | Required or preferred |
Mark the difference between a functional requirement and an aesthetic preference. “Must show both faces” affects use of a double-faced card. “Prefer borderless art” does not. If the requested treatment is unavailable, that distinction tells you whether an ordinary readable version is acceptable or whether the card needs review before fulfillment.
Visual cohesion can make a deck easier to scan, but consistency should not erase useful distinctions. A commander should be immediately recognizable, lands should read as lands, and commonly confused interaction pieces should not all use nearly identical text-light treatments. If you want a unified presentation, the guide to cohesive Commander deck themes offers a gameplay-first way to choose frames and art.
Step 4: Plan double-faced cards and gameplay extras
Double-faced cards create a physical handling decision. A represented card needs enough information for both faces to be identified and played correctly. Depending on the ordering option and your sleeve setup, that could mean one double-faced piece, a front-facing deck card plus a clearly labeled reference piece, or another transparent arrangement accepted by your group. Decide before ordering rather than improvising after the cards arrive.
Build the helper package from what the deck actually produces. Search the locked list for words and mechanics that create tokens, emblems, copies, predefined dungeons or similar external objects. Then distinguish recurring pieces from remote edge cases. A creature-token deck may need several copies of its common token, while a single emblem can usually be represented by one clear card.
- Tokens the deck creates regularly, with distinct power, toughness, colors, and abilities visible
- Emblems or other named game objects required by specific cards
- Day/night, monarch, initiative, storm, poison, experience, or commander-damage trackers when relevant
- Clearly labeled copy tokens for cloned permanents or spells
- Reference pieces for double-faced cards when your chosen physical setup needs them
- Dice or counters for repeated game-state information
These pieces should not masquerade as members of the core deck. Store them in a different sleeve color, a labeled divider section, or a separate token box. That prevents a copy token from joining the library during cleanup, where it will inevitably wait until the worst possible opening hand to announce itself.
Step 5: Keep replacement slots outside the deck
A cEDH list is rarely frozen forever. New releases, local metagame changes, and testing results can all create reasonable swaps. The mistake is placing several alternatives in the deck box without recording which card each one replaces.
Create a replacement register with three fields: incoming card, outgoing card, and reason for the test. Sleeve optional cards differently or keep them behind a divider. Before each session, apply the intended changes to a new decklist revision and verify that the active deck still totals 100 cards.
This preserves the original build while allowing deliberate iteration. It also stops an experimental card from becoming a permanent 101st card through the ancient deckbuilding process known as “I am pretty sure I took something out.” Strategic questions about whether a swap improves a particular matchup belong with a dedicated cEDH strategy resource such as MTGEDH; the physical-deck task is to record the decision accurately.
Step 6: Reconcile the completed deck before play
Do not begin with shuffling. Lay out the command-zone cards, sort the remaining cards by name, and compare them line by line with the locked export. Counting to 100 catches missing pieces but cannot catch a wrong card replacing the right one.
- Confirm the correct commander or commanders.
- Count the command zone and library together to 100.
- Compare every card name and quantity with the locked file.
- Verify basic-land quantities rather than checking only the total land count.
- Inspect requested versions and separate requirements from merely preferred treatments.
- Confirm that both faces or references for double-faced cards are present and readable.
- Move all tokens, trackers, and replacement cards outside the core deck.
- Check card names, mana costs, type lines, and rules text for table readability.
- Save the reconciled list as the physical deck’s current revision.
A general proxy deck arrival checklist can help with physical and fulfillment checks after the list audit. Commander Proxies also describes its own process as using HP Indigo printing, UV coating, and die cutting, but those are first-party production descriptions rather than independent performance findings. List accuracy and physical inspection remain separate checks.
Readability matters more when decisions are fast
Competitive games can involve dense boards, layered interaction, and short decision windows. A card treatment that requires repeated explanation creates avoidable friction even when its rules identity is technically known. Prioritize visible names, mana costs, type lines, rules information, and recognizable distinctions between similarly functioning cards.
Alternate art is easiest to use when it preserves those anchors. Textless or heavily altered treatments deserve extra caution for interaction, combo pieces, and cards opponents must evaluate immediately. Cohesion is good; turning the battlefield into a memory quiz is less good.
Confirm proxy and power expectations separately
Proxy acceptance and deck power are two different conversations. A group might accept a fully proxied deck but not expect cEDH gameplay, or it might run high-powered games while imposing specific requirements on represented cards. State both facts plainly: the deck is proxied, and it is intended for competitive Commander play.
For a store or organized event, ask the organizer before arriving and check the current governing documents rather than relying on a previous event or community folklore. Wizards maintains current material through its WPN Rules and Documentation page and a dedicated WPN Policy Hub. Requirements can depend on the event and organizer, so this article cannot establish whether a fully proxied deck is permitted at a particular table.
For ordinary Commander expectations, Wizards presents Brackets and Game Changers as pregame communication tools alongside the format rules. They can help start a power conversation, but a direct “this is a cEDH list” remains clearer than asking a broad label to do all the work. The guide to building a proxy deck that fits your playgroup covers that expectation-setting process for non-cEDH tables.
The final rule: change the list deliberately
A reliable full deck is not the one with the fanciest treatment. It is the one you can trace back to a named, locked list. Keep the legal 100 separate from helpers, document requested versions, label every test swap, and reconcile the physical cards against the export before game night.
If you are still deciding which win package or interaction suite belongs in the deck, solve that strategy question first through a cEDH-focused resource such as MTGEDH. If you intend to print and cut the cards at home, use a dedicated DIY resource such as Let’s Proxy. Once the list itself is settled, freeze a revision and run the fidelity workflow from commander to card 100.