Your decks make PAN better.
PAN is an inference tool, which means unusual decks sometimes expose places where it sees the cards but misses the reason they belong together. We review anonymous analysis results, find weak reads, and improve the analyzer. This is where we show what changed.
September 18, 2026Deck Lab editing, Commander Contracts, usage counters & project cleanupView updates
Card Audition became a real editing loop
Card Audition now separates Structural Fit from the effect independently measured by Commander PAN, uses legal card autocomplete, and tests conservative one-card swaps against the finished deck. Applying a swap now recalculates the modified 99-card library instead of leaving coverage and validation tied to the original build.
Applied cards also receive a freshly evaluated Primary Job rather than inheriting the removed card's board slot. Commander Role remains separate, so a card can satisfy a commander's requirement without losing its actual functional job.
Commander Contracts expanded beyond creature types
Deck Lab's hard-requirement layer now recognizes reusable rules-text-derived functional requirements, including creature-subtype entry, friendly single-target spells, non-subtype creature entry, and actual -1/-1 counter placement. The same contract semantics are used by deck construction, Commander Roles, Card Audition, and applied-swap recalculation.
Deck Lab and Pauper gained their own usage counters
Deck Lab now tracks successful builds separately, while Pauper exposes its own successful-analysis count. Deck Lab builds do not inflate the existing site-wide Commander + Pauper analysis counter. The example commander-name chips beneath Deck Lab search were also removed for a cleaner starting screen.
Project documentation was consolidated
The growing collection of release-specific notes was consolidated into a current project README, technical architecture reference, and changelog. Historical implementation notes remain archived without cluttering the project root.
September 17, 2026Strategy architecture, validation & workflowView updates
Shared Strategy Vocabulary
Deck Lab and regular Commander PAN now share an expanding strategy vocabulary without sharing answers. Regular PAN independently evaluates the submitted 100 cards rather than inheriting Deck Lab's Strategy Bridge or assumptions. Roll Dice and Graveyard Spell Replay were the first bridge-native concepts promoted into this shared recognition layer.
Smarter Strategy Bridges
Strategy Bridges now pay closer attention to causality: what the command zone produces versus what the rest of the deck must actually provide. +1/+1-counter strategies require broad counter distribution or explicit counter payoffs rather than incidental self-counters or copy riders. Generalized causal bridges also support dice-driven commander growth and graveyard instant/sorcery combat replay.
Closure Quality & Build Integrity
Deck Lab now distinguishes cards that operate an engine from cards that actually help finish the game. Copied-spell combat strategies can recognize pump, trample, and double strike as genuine closure. Build Integrity also checks whether the strategic architecture is complete instead of treating any filled 99-card library as automatically finished.
Paired Commander Validation
Partner, Background, and other legal two-card command zones are preserved when completed Deck Lab builds are sent to regular Commander PAN for independent validation, restoring PAN validation and Strategy Repair for paired command zones.
UI & state cleanup
Copy full decklist moved beside the TCGplayer action, a Clear control was added, and starting a new build now resets paired-command-zone state so an old secondary commander cannot leak into the next build.
September 14, 2026Deck Lab foundation: Strategy Bridge, Repair, Candidate Quality & interface fixesView updates
Strategy Bridge
Deck Lab can derive a provisional deck architecture from the command zone before the 99 exists. The first generalized bridge connected permanent-copy commanders to the same Creature Copy / Tokens idea regular PAN can independently recognize in finished decks.
The bridge is rules-text driven rather than commander-name specific. Contribution Accounting still checks whether the finished library actually supplies the jobs the strategy claims to need.
Strategy Repair
Deck Lab sends its first assembled draft through regular Commander PAN as an independent critique. When the observed deck drifts from the command-zone Strategy Bridge, PAN can make a small set of conservative swaps, preserve coverage and connected packages, re-analyze the repaired deck, and keep the repair only when Commander PAN validates the result.
Candidate Quality
Qualified cards are compared on efficiency, reliability, repeatability, Commander adoption, package coverage, and role-specific quality instead of treating every semantic match as equally good. Strategy fit and coverage still lead assembly, while Commander PAN remains the validation guard against strategic drift.
Additional strategy support
Generalized Graveyard / Permanent Recursion and -1/-1 Counters / Tokens packages were added from command-zone rules and semantics. Strategy Repair was also taught not to trade away an already-correct strategic identity for a superficially higher score.
Interface and stability fixes
Deck Lab gained PAN's shared header/navigation, commander art, board hover previews, and a TCGplayer mass-entry action. Direct Scryfall image URLs were restored after an image-proxy regression, and a minimal animated busy indicator replaced a mana-symbol experiment that caused a browser-side freeze.
September 8, 2026Better commander synergy: Auras, Voltron & damage amplificationView updates
PAN improved how it connects a commander's rules text to dense packages in the 99 when the commander is the reason those cards function as a coherent engine.
Aura & Voltron decks
Aura-heavy decks can now receive more weight for commander-mediated Aura relationships and recognize a concentrated Voltron plan and commander-damage finish when the deck supports it.
Damage-amplification decks
PAN can better distinguish a spellslinger shell from a commander that converts repeated small damage events into the deck's defining pressure engine.
These changes are relationship-gated. Cards in the deck still have to meaningfully support the commander's text.
September 2026Pauper joins PAN & Commander parsing improvesView updates
Pauper analysis joins PanEDH.com
PAN Flute extends PAN's Pressure, Access, and Networking approach to Pauper. It looks for relationships between engines, payoffs, interaction, and the deck's clearest route to winning rather than forcing every list into a stock archetype label.
More reliable commander selection and deck parsing
Commander analysis better handles lists containing multiple legendary creatures, alternate printings and reskins, double-faced card names, quantities, set annotations, and other common deck-export formatting. When PAN finds more than one plausible commander, players can select the correct one before the report is generated.