# UltraCompressed Docs Mode
# ~70% token reduction via telegram-style+symbols+abbrevs

Activation:
  Flags: --ultracompressed | --uc
  Natural: "ultracompressed" | "minimal tokens" | "telegram style"
  Auto: Context>70% | Token budget specified | User preference

Core_Rules:
  Remove_Words:
    Articles: the|a|an
    Conjunctions: and|or|but|however|therefore
    Prepositions: of|in|on|at|to|for|with|from|by
    Fillers: that|which|who|very|really|quite|just
    Verbose: "in order to"→to | "make sure"→ensure | "as well as"→&
  
  Symbol_Map:
    →: to/leads to/results in
    &: and/with/plus
    @: at/located at
    w/: with/including
    w/o: without/excluding
    +: add/plus/include
    -: remove/minus/exclude
    ∴: therefore/thus
    ∵: because/since
    ≈: approximately/about
    ∀: for all/every
    ∃: exists/there is
    ⊂: subset of/part of
    ≡: equivalent to/same as

  Abbreviations:
    cfg: configuration      fn: function
    impl: implementation    req: require/request
    resp: response         auth: authentication
    db: database           api: API/interface
    env: environment       dev: development
    prod: production       deps: dependencies
    arg: argument          param: parameter
    val: value            obj: object
    arr: array            str: string
    num: number           bool: boolean
    err: error            msg: message
    usr: user             sys: system
    lib: library          pkg: package
    ctx: context          ref: reference
    docs: documentation    spec: specification

Structure_Rules:
  Headings: <20 chars | No articles | Symbols OK
  Sentences: <50 chars | Telegram style | No punctuation if clear
  Paragraphs: <100 chars | 3 sentences max | Lists preferred
  Lists: Bullets>numbers | No full sentences | Key info only
  Tables: Headers abbreviated | Cell content minimal | Symbols OK

Format_Hierarchy:
  1. YAML/JSON (structured data)
  2. Tables (comparison/reference)
  3. Bullet lists (enumeration)
  4. Numbered lists (sequences)
  5. Prose (avoid when possible)

Content_Rules:
  NO: Introductions|Conclusions|Transitions|Explanations of obvious
  NO: "Getting Started"|"Overview"|"Introduction" sections
  NO: Filler phrases|Politeness|Redundancy
  YES: Direct information|Code>text|Examples>description
  YES: Symbols>words|Abbreviations>full terms|Active>passive

Legend_Generation:
  When: Start of each doc | When new symbols/abbrevs used
  Format: Compact table | Only used items | Sort by frequency
  Location: After title | Before content | Collapsible if supported
  Example: |
    ## Legend
    | Symbol | Meaning | | Abbrev | Meaning |
    |--------|---------|---|--------|---------|
    | → | leads to | | cfg | configuration |
    | & | and/with | | fn | function |
    | w/ | with | | impl | implementation |

Example_Transformations:
  Before: "This section provides an introduction to getting started with the authentication system"
  After: "Auth Setup"
  
  Before: "In order to configure the database connection, you need to set the following environment variables"
  After: "DB cfg: Set env vars:"
  
  Before: "The function takes three parameters and returns a boolean value"
  After: "fn(3 params)→bool"

Quality_Checks:
  Token_Reduction: Target 70% | Measure before/after
  Clarity_Score: Must remain >80% | Test w/ context
  Info_Density: Maximize bits/token | No info loss
  Legend_Required: Always include for symbols & abbreviations