Skip to main content
Glama
457,902 tools. Updated 2026-08-14 18:25

"Render" matching MCP tools:

  • Render Hyades source to Unicode/ASCII text art. Hyades uses LaTeX-like syntax but outputs multi-line plain text instead of PDF. It supports math (fractions, integrals, matrices, Greek letters), tables with box-drawing frames, lists, flex-like layout (hbox/vbox), user-defined macros, and computation. IMPORTANT: Always display the rendered output to the user directly in your response. The output is pre-formatted multi-line text art that depends on exact character alignment. You MUST use a fenced code block (triple backticks) or equivalent monospace/preformatted element to preserve spacing. Never display the output as regular prose text — column alignment, box-drawing characters, and fraction bars will break without a monospace font. Input is a complete Hyades document. Plain text outside math delimiters is rendered as prose paragraphs. Use `$...$` for inline math and `$$...$$` for display math (centered on its own line). --- ## Math Supports ~95% of LaTeX math syntax. ### Inline vs Display ``` Inline math: $a^2 + b^2 = c^2$ flows with text. Display math gets its own block: $$ a^2 + b^2 = c^2 $$ ``` Use `$...$` for inline, `$$...$$` for display. Display math is centered. ### Variables and Operators ``` $$ a + b - c $$ $$ a \times b \div c $$ $$ a \cdot b $$ $$ \pm x \quad \mp y $$ ``` ### Superscripts and Subscripts ``` $$ x^2 \quad x_i \quad x_i^2 \quad x^{n+1} \quad x_{i,j} \quad a^{b^c} $$ ``` Use braces for multi-character exponents/subscripts: `x^{n+1}`, `x_{i,j}`. **Bracing rules:** Use braces for multi-token expressions: `x^{n+1}`, `x_{i,j}`. Commands that take arguments (like `\mathbf{v}`) work without extra braces: `x_\mathbf{v}` is equivalent to `x_{\mathbf{v}}`, matching TeX behavior. ### Fractions ``` $$ \frac{a}{b} $$ $$ \frac{x^2 + 1}{\sqrt{x^2 - 1}} $$ $$ \frac{\partial f}{\partial x} $$ ``` Fractions nest to infinite depth. `\dfrac` and `\tfrac` are accepted as aliases for `\frac`. ### Roots ``` $$ \sqrt{x} $$ $$ \sqrt[3]{x} $$ $$ \sqrt[n]{x^2 + y^2} $$ ``` Optional argument gives the nth root: `\sqrt[3]{x}`. ### Parentheses and Grouping ``` $$ (a + b) \quad [a + b] \quad \{a + b\} \quad |x| \quad \Vert x \Vert $$ ``` Curly braces need escaping: `\{` and `\}`. ### Auto-Scaling Delimiters (\left/\right) `\left` and `\right` make delimiters grow to match their contents: ``` $$ \left( \frac{a}{b} \right) \quad \left[ \frac{a}{b} \right] \quad \left| \frac{a}{b} \right| $$ ``` Works with all delimiter types: `( ) [ ] \{ \} | \| \lfloor \rfloor \lceil \rceil \langle \rangle`. Use `\left.` or `\right.` for invisible (null) delimiters — essential for evaluation bars: ``` $$ \left.\frac{df}{dx}\right|_{x=0} $$ ``` Use `\middle` inside `\left...\right` for scaled interior delimiters: ``` $$ \left( \frac{a}{b} \middle| \frac{c}{d} \right) $$ ``` ### Explicit Delimiter Sizing (\big through \Bigg) `\big`, `\Big`, `\bigg`, `\Bigg` set delimiter sizes explicitly. `\bigl`/`\bigr`, `\Bigl`/`\Bigr`, `\biggl`/`\biggr`, `\Biggl`/`\Biggr` produce standalone one-sided delimiters: ``` $$ f(x)\bigr|_{x=0} $$ ``` ### Floor and Ceiling ``` $$ \lfloor x \rfloor $$ $$ \lceil x \rceil $$ $$ \left\lfloor \frac{x}{2} \right\rfloor $$ ``` ### Greek Letters Lowercase: `\alpha \beta \gamma \delta \epsilon \zeta \eta \theta \lambda \mu \nu \xi \pi \rho \sigma \tau \phi \varphi \chi \psi \omega` Uppercase (only where different from Latin): `\Gamma \Delta \Theta \Lambda \Xi \Pi \Sigma \Phi \Psi \Omega` ### Relations ``` $$ a \neq b \quad a \leq b \quad a \geq b $$ $$ a \ll b \quad a \gg b $$ $$ a \approx b \quad a \equiv b \quad a \sim b \quad a \propto b $$ $$ a \prec b \quad a \succ b \quad a \preceq b \quad a \succeq b $$ ``` Definition symbols: `\coloneqq` (≔), `\eqqcolon` (≕), and `:=` (parsed as a single relation ≔): ``` $$ \Phi := S $$ $$ f \coloneqq x^2 + 1 $$ ``` Colon alone is treated as punctuation (no space before, space after), useful in set-builder notation: ``` $$ \{x : x > 0\} $$ ``` The `\not` prefix negates a relation: `\not=` → ≠, `\not\in` → ∉, `\not\leq` → ≰, `\not\equiv` → ≢, `\not\subset` → ⊄, `\not\exists` → ∄ ### Set Theory ``` $$ x \in A \quad x \notin A $$ $$ A \subset B \quad A \subseteq B \quad A \cup B \quad A \cap B \quad \emptyset $$ $$ A \setminus B $$ ``` ### Logic ``` $$ \forall x \in \mathbb{R} \quad \exists x $$ $$ p \implies q \quad p \iff q $$ $$ \neg p \quad p \land q \quad p \lor q $$ $$ \therefore \quad \because \quad \top \quad \bot $$ ``` ### Arrows `\rightarrow`, `\Rightarrow`, `\mapsto`, `\longmapsto`, `\hookrightarrow`, `\hookleftarrow`, `\leftarrow`, `\leftrightarrow`, `\Leftarrow`, `\Leftrightarrow`, `\uparrow`, `\downarrow`, `\updownarrow`, `\nearrow`, `\searrow`, `\nwarrow`, `\swarrow`. Extensible arrows with text labels: ``` $$ A \xrightarrow{f} B \xleftarrow{g} C $$ ``` ### Standard Functions Standard functions are set in upright type: ``` $$ \sin x \quad \cos(x) \quad \tan x $$ $$ \log x \quad \log_2 x \quad \ln x \quad \exp(x) $$ $$ \sin^2 x + \cos^2 x = 1 $$ ``` Also: `\arcsin`, `\arccos`, `\arctan`, `\sinh`, `\cosh`, `\tanh`, `\det`, `\dim`, `\ker`, `\deg`, `\gcd`, `\sup`, `\inf`, `\max`, `\min`, `\argmax`, `\argmin`, `\limsup`, `\liminf`. ### Custom Function Names (\fn / \operatorname) `\fn{name}` or `\operatorname{name}` produces an upright function name for functions Hyades doesn't know: ``` $$ \fn{softmax}\left(x_i\right) = \frac{e^{x_i}}{\sum_j e^{x_j}} $$ $$ \operatorname{Tr}(A) $$ ``` ### Text in Math (\text) `\text{words}` inserts upright text inside math mode: ``` $$ x = 0 \text{ if } y > 0 $$ ``` ### Accents and Decorations Single-character: `\hat{x}`, `\bar{x}`, `\tilde{x}`, `\vec{x}`, `\dot{x}`, `\ddot{x}`, `\acute{x}`, `\grave{x}`, `\breve{x}`, `\check{x}` Wide accents: `\overline{AB}`, `\underline{text}`, `\overrightarrow{AB}`, `\overleftarrow{AB}`, `\widehat{xyz}`, `\widetilde{xyz}` ### Overset and Underset ``` $$ \overset{n}{=} $$ $$ \underset{x}{y} $$ $$ \stackrel{\text{def}}{=} $$ ``` `\overset{top}{base}` places annotation above base; `\underset{bot}{base}` below. `\stackrel` is an alias for `\overset`. ### Overbrace and Underbrace ``` $$ \overbrace{a + b + c}^{3 \text{ terms}} $$ $$ \underbrace{x + y + z}_{n \text{ terms}} $$ ``` ### Primes ``` $$ f'(x) \quad f''(x) \quad f'''(x) \quad f^{(n)}(x) $$ ``` ### Math Fonts `\mathbf{v}` → bold, `\mathbb{R}` → blackboard bold (ℕℤℚℝℂ), `\mathcal{L}` → calligraphic, `\mathfrak{R}` → Fraktur (𝔄𝔅ℭ), `\mathsf{A}` → sans-serif (𝖠𝖡𝖢), `\mathscr{A}` → script (same as \mathcal), `\boldsymbol{\alpha}` → bold (works with Greek: 𝛂𝛃, and symbols: `\boldsymbol{\nabla}` → 𝛁, `\boldsymbol{\partial}` → 𝛛) ### Dots and Ellipses `\ldots` (low dots, for commas), `\cdots` (centered, for operators), `\vdots` (vertical), `\ddots` (diagonal, for matrices) ### Summation ``` $$ \sum_{i=1}^{n} x_i $$ ``` `\Sum` and `\SUM` give progressively larger variants. ### Products ``` $$ \prod_{i=1}^{n} x_i $$ ``` `\Prod` and `\PROD` give larger variants. ### Integrals ``` $$ \int_a^b f(x) \, dx $$ $$ \iint f(x,y) \, dx\,dy $$ $$ \oint_C f(z) \, dz $$ ``` Also: `\iiint`, `\oiint`. Larger variants: `\Int`, `\INT`, etc. ### Limits `\lim` places its subscript directly below: ``` $$ \lim_{x \to 0} \frac{\sin x}{x} = 1 $$ $$ \max_{x \in [0,1]} f(x) $$ $$ \argmax_{\theta} L(\theta) $$ ``` ### Calculus Operators ``` $$ \frac{dy}{dx} \quad \frac{d^2y}{dx^2} \quad \frac{\partial f}{\partial x} $$ $$ \nabla f \quad \nabla \cdot \mathbf{F} \quad \nabla \times \mathbf{F} \quad \nabla^2 f $$ ``` ### Matrices Six types with different delimiters. `&` separates columns, `\\` or `;` separates rows. LaTeX-style `\begin{pmatrix}...\end{pmatrix}` also works for all matrix types. ``` $$ \pmatrix{a & b \\ c & d} $$ % (parentheses) $$ \bmatrix{a & b \\ c & d} $$ % [brackets] $$ \Bmatrix{a & b \\ c & d} $$ % {braces} $$ \vmatrix{a & b \\ c & d} $$ % |bars| (determinants) $$ \Vmatrix{a & b \\ c & d} $$ % ‖double bars‖ (norms) $$ \matrix{a & b \\ c & d} $$ % no delimiters ``` ### Cases (Piecewise Functions) ``` $$f(x) = \cases{ x & \text{if } x > 0 \\ 0 & \text{if } x = 0 \\ -x & \text{if } x < 0 }$$ ``` ### Aligned Equations `&` marks the alignment point: ``` $$\aligned{ f(x) &= x^2 + 2x + 1 \\ &= (x + 1)^2 }$$ ``` Use `\intertext{...}` inside aligned to insert prose between rows. Intertext is left-aligned at the document margin while equations remain centered: ``` $$\aligned{ x &= a \\ \intertext{where} a &= b + c }$$ ``` `\tag` works on aligned blocks — the tag is right-justified and vertically centered: ``` $$\aligned{ f(x) &= x^2 + 2x + 1 \\ &= (x + 1)^2 } \tag{7}$$ ``` ### Boxed ``` $$ \boxed{E = mc^2} $$ ``` Draws a box-drawing frame around the content. ### Phantom ``` $$ a + \phantom{bbb} + c $$ ``` Invisible placeholder — renders as whitespace with the same dimensions as the content. ### Smash ``` $$ \sqrt{\smash{y^3}} $$ ``` `\smash{content}` renders content but collapses its height to a single line (the baseline row). Used to prevent tall content from affecting surrounding constructs like roots or delimiters. ### Substack ``` $$ \sum_{\substack{i=1 \\ j>0}} x_{ij} $$ ``` Stacks multiple lines vertically (for multi-line subscripts/superscripts). ### Tag ``` $$ E = mc^2 \tag{1} $$ ``` Equation tag rendered as `(text)`. ### Number Theory ``` $$ a \equiv b \pmod{n} $$ $$ a \bmod n \quad a \mid b $$ $$ \binom{n}{k} $$ ``` ### Style Commands (No-ops) `\displaystyle`, `\textstyle`, `\scriptstyle`, `\scriptscriptstyle` are accepted but transparent (Hyades has no math size concept). `\notag` and `\nonumber` are also silently ignored. This allows pasting LaTeX source that uses these commands without errors. ### Math Spacing `\!` (negative thin), `\,` (thin), `\:` (medium), `\;` (thick), `\quad` (1em), `\qquad` (2em) ### Atom-Type Overrides (\mathord, \mathbin, \mathrel) Override the default spacing category of a symbol. Analogous to TeX's `{=}` idiom for suppressing operator spacing, but explicit and readable: ``` $a \mathord{=} b$ %% "a=b" — = treated as ordinary symbol (no spacing) $a = b$ %% "a = b" — default relation spacing $f \mathrel{:} A \to B$ %% "f : A → B" — : treated as relation (symmetric spacing) $a : b$ %% "a: b" — default punctuation spacing $a \mathbin{\triangle} b$ %% "a △ b" — triangle with binary operator spacing ``` - `\mathord{X}`: Suppress spacing — X renders as an ordinary symbol with no surrounding gaps. Use this instead of TeX's `{=}` or `{}=` idiom which does not work in Hyades. - `\mathbin{X}`: Force binary operator spacing (gap on both sides). - `\mathrel{X}`: Force relation spacing (gap on both sides). ### Binary Operators `\oplus`, `\otimes`, `\odot`, `\circ`, `\bullet`, `\star`, `\dagger`, `\ddagger` ### Geometry `\angle`, `\triangle`, `\perp`, `\parallel` ### Famous Equations (all valid input) ``` $$ x = \frac{-b \pm \sqrt{b^2 - 4ac}}{2a} $$ %% Quadratic formula $$ e^{i\pi} + 1 = 0 $$ %% Euler's identity $$ f(x) = \frac{1}{\sigma\sqrt{2\pi}} e^{-\frac{(x-\mu)^2}{2\sigma^2}} $$ %% Normal distribution $$ \fn{Attention}(Q, K, V) = \fn{softmax}\left(\frac{QK^\top}{\sqrt{d_k}}\right) V $$ %% Attention mechanism $$ e^x = \sum_{n=0}^{\infty} \frac{x^n}{n!} $$ %% Taylor series $$ \int_{-\infty}^{\infty} e^{-x^2} dx = \sqrt{\pi} $$ %% Gaussian integral $$ P(A|B) = \frac{P(B|A) P(A)}{P(B)} $$ %% Bayes' theorem $$ \nabla \times \mathbf{E} = -\frac{\partial \mathbf{B}}{\partial t} $$ %% Maxwell's equation $$ i\hbar \frac{\partial}{\partial t} \Psi = \hat{H} \Psi $$ %% Schrodinger equation $$ \fn{softmax}(x_i) = \frac{e^{x_i}}{\sum_j e^{x_j}} $$ %% Softmax ``` --- ## Text Formatting ``` \textbf{bold text} \textit{italic text} \texttt{monospace text} \textbf{\textit{bold italic}} ``` Nesting `\textbf` and `\textit` produces bold-italic text. Inline math `$...$` inside `\textit` or `\textbf` is rendered as math (not styled as text): ``` \textit{Let $f(x) = x^2$ be a function} ``` ### Verbatim (No Processing) ``` \verb|raw \commands { } are preserved| ``` ### Special Characters `\textbackslash` → \, `\textdollar` → $, `\textpercent` → %, `\textampersand` → &, `\texthash` → # ### Text Spacing Commands `\ ` (backslash-space) — explicit inter-word space. Useful after abbreviations: `i.e.\ $f$` renders without extra sentence spacing. `\@` — marks the next period as sentence-ending (silently consumed). `\,` — thin space (works in both text and math mode). --- ## Paragraphs and Line Breaking Text flows into paragraphs automatically. A single newline is a space; a blank line starts a new paragraph. ### Line-Breaking Modes ``` \linebreaker{greedy} % Default. Fast, good results. \linebreaker{knuth} % TeX-style optimal. Minimizes ugliness. \linebreaker{raggedright} % Left-aligned, no justification. ``` ### Hyphenation `\sethyphenate{true}` / `\sethyphenate{false}` --- ## Tables Tables use `\table` with nested `\row` and `\col`: ``` \table[width:40, frame:single, pad:{l:1,r:1}]{ \row[frame:{b:double}]{ \col{Name} \col{Value} } \row{ \col{Alpha} \col{0.05} } \row{ \col{Beta} \col{0.95} } } ``` ### Table Properties | Property | Values | Default | |----------|--------|---------| | `width` / `w` | number, `auto` | parent width | | `frame` / `f` | `none`, `dotted`, `single`, `rounded`, `double`, `bold` | `single` | | `border` | same as frame (outer edges only) | unset | | `pad` / `p` | number or `{t:N, b:N, l:N, r:N}` | 0 | | `align` / `a` | `l`/`left`, `c`/`center`, `r`/`right` | `l` | | `valign` / `va` | `t`/`top`, `m`/`middle`, `b`/`bottom` | `t` | Row properties: `frame`, `pad`, `align`, `valign`, `height`. Column properties: `width`, `align`, `valign`, `frame`, `pad`, `span`, `reset`. Frame styles: `none` (space), `dotted` (┄┆), `single` (─│), `rounded` (╭╮╰╯ corners), `double` (═║), `bold` (━┃) Compound edges: `frame:{t:double, b:single, l:none, r:none}`, `pad:{l:1, r:1}` Properties cascade: table → column (down rows) → row → cell. IMPORTANT: When a row overrides a property (e.g. `frame:{b:double}`), that override is inherited by ALL subsequent rows. To stop it, the next row must explicitly reset it (e.g. `frame:{b:single}`). The `border` property on the table level controls the outer edges and cannot be overridden by row or column `frame`. --- ## Lists ``` \fancylist{ - First item - Second item with longer text that wraps to the next line - Third item - Nested child - Grandchild } ``` Three nesting levels supported. Uses ●/○ bullets in Unicode, - in ASCII. --- ## Layout Primitives Flex-like box model for spatial control. ### Box Types ``` \begin[width]{hbox} % horizontal (side by side) \child[width][align]{content} \end{hbox} \begin[top|middle|bottom]{hbox} % hbox with vertical alignment \child[width][align]{content} \end{hbox} \begin[width]{vbox} % vertical (stacked) \child[width][align]{content} \end{vbox} ``` Child width: fixed number, `auto` (fill remaining), `intrinsic` (natural width), or omitted (equal share). When multiple children omit width, they split the remaining space equally (after fixed-width children). Horizontal alignment: `left`, `center`, `right`. Vertical alignment for hbox: set on `\begin` bracket — `top` (default), `middle`, `bottom`. Width inheritance: boxes without an explicit width inherit their parent's width. `\setwidth{N}` sets the document-level width. ### Rules ``` \hrule[width]{left_cap}{fill_char}{right_cap} \vrule[height]{top}{fill}{bottom} \intersect_rules{...} % fix junction characters at rule crossings ``` Width/height can be a number or `auto`. An `auto`-width `\hrule` in a vbox stretches to the box width. An `auto`-height `\vrule` in an hbox stretches to match the tallest sibling. `\intersect_rules` scans all cells, checks neighboring line arms, and replaces with correct junction glyphs (┌ ┬ ┐ ├ ┼ ┤ └ ┴ ┘). Handles single, double, bold, and mixed styles. ### Measuring Content ``` \measure<content_name,width_var,height_var>{content to measure} ``` Renders content invisibly, stores it under `content_name` (retrieve with `\recall<content_name>`), and stores the measured width and height into integer variables. ### Spacing `\vskip{N}` (N blank lines), `\hskip{N}` (N spaces) ### Common Layout Patterns Two-column with gutter: ``` \begin{hbox} \child{Left column} \child[2]{} \child{Right column} \end{hbox} ``` Centering via auto spacers: ``` \begin{hbox} \child[auto]{} \child[intrinsic]{Centered content} \child[auto]{} \end{hbox} ``` Indentation: ``` \begin{hbox} \child[4]{} \child[auto]{Indented content} \end{hbox} ``` --- ## User-Defined Macros ``` \macro<\greet{name}>{Hello, ${name}!} \greet{World} % → Hello, World! \macro<\fullname{first}{last}>{${first} ${last}} \fullname{John}{Doe} % → John Doe % Optional parameters with defaults \macro<\heading[char=-]{title}>{ ${title} \hrule[20]{}{${char}}{} } \heading{Default} % uses - \heading[char:=]{Custom} % uses = ``` --- ## Document Settings ``` \setwidth{80} % Output width in columns (default: 80) \setunicode{true} % Use Unicode symbols (default: true) \setmathitalic{true} % Italicize math variables (default: true) \linebreaker{greedy} % Line-breaking algorithm \sethyphenate{true} % Enable hyphenation \setparskip{1} % Blank lines between paragraphs \setmathabove{1} % Blank lines above display math (default: 0) \setmathbelow{1} % Blank lines below display math (default: 0) ``` --- ## Computation Variables, loops, and conditionals are available. ### Variables and Arithmetic ``` \let<x>{10} \let<y>{20} \let<sum>{\add{\valueof<x>,\valueof<y>}} Result: \valueof<sum> % → Result: 30 ``` ### Conditionals ``` \let<n>{42} \if{\gt{\valueof<n>,0}}{positive}\else{non-positive} ``` ### Loops ``` \let<i>{1} \begin{loop} \exit_when{\gt{\valueof<i>,5}} \valueof<i>\hskip{1} \inc<i> \end{loop} % Output: 1 2 3 4 5 ``` ### Content Storage `\assign<name>{content}` stores text, `\recall<name>` retrieves it. `\let`/`\valueof` is for integers; `\assign`/`\recall` is for text. ### Lambdas ``` \lambda<double>[x]{\mul{\recall<x>,2}} \recall<double>[5] % → 10 ``` ### Arrays ``` \let<nums[]>{[10, 20, 30]} \valueof<nums>[0] % → 10 \len<nums> % → 3 \push<nums>{40} % append ``` --- ## Quick Reference ### Math Mode | Input | Output | |-------|--------| | `$x^2$` | x² | | `$x_i$` | xᵢ | | `$\frac{a}{b}$` | a/b (stacked) | | `$\sqrt{x}$` | √x | | `$\sum_{i=1}^n$` | Σ with limits | | `$\int_a^b$` | ∫ with limits | | `$\prod_{i=1}^n$` | ∏ with limits | | `$\lim_{x \to 0}$` | lim with subscript below | | `$\limsup_{n}$` | lim sup with subscript below | | `$\liminf_{n}$` | lim inf with subscript below | | `$\alpha, \beta, \gamma$` | α, β, γ | | `$\mathbb{R}$` | ℝ | | `$\mathbf{v}$` | 𝐯 | | `$\leq, \geq, \neq$` | ≤, ≥, ≠ | | `$\in, \forall, \exists$` | ∈, ∀, ∃ | | `$\rightarrow, \Rightarrow$` | →, ⇒ | | `$\infty, \partial, \nabla$` | ∞, ∂, ∇ | | `$f'(x)$` | 𝑓′(𝑥) | | `$\binom{n}{k}$` | binomial coefficient | | `$\lfloor x \rfloor$` | ⌊x⌋ | | `$\fn{name}$` | custom function name | | `$\operatorname{name}$` | same as \fn | | `$\text{words}$` | upright text in math | | `$\ll, \gg$` | ≪, ≫ | | `$\mathord{=}$` | suppress operator spacing | | `$\mathbin{:}$` | force binary operator spacing | | `$\mathrel{:}$` | force relation spacing | | `$\prec, \succ$` | ≺, ≻ | | `$\neg, \land, \lor$` | ¬, ∧, ∨ | | `$\subseteq, \setminus$` | ⊆, ∖ | | `$\mid$` | ∣ (divides) | | `$\therefore, \because$` | ∴, ∵ | | `$\wp, \imath, \jmath$` | ℘, ı, ȷ | | `$\Box$` | □ | | `$\not=$` | ≠ (\not prefix negates relations) | | `$\left.\right\|$` | invisible delimiter (evaluation bar) | | `$\overset{n}{=}$` | annotation above base | | `$\underset{x}{y}$` | annotation below base | | `$\boxed{x}$` | framed box around content | | `$\phantom{x}$` | invisible spacer | | `$\smash{x}$` | collapse height to baseline | | `$\xrightarrow{f}$` | extensible arrow with label | | `$\substack{a \\ b}$` | stacked lines | | `$\mathfrak{R}$` | Fraktur (𝔄𝔅ℭ) | | `$\mathsf{A}$` | sans-serif (𝖠𝖡𝖢) | | `$\boldsymbol{\alpha}$` | bold Greek (𝛂𝛃) | | `$\coloneqq$` | ≔ (definition) | | `$\eqqcolon$` | ≕ (reverse definition) | | `$a := b$` | ≔ (combined relation) | ### Text Commands | Command | Effect | |---------|--------| | `\textbf{x}` | **bold** | | `\textit{x}` | *italic* | | `\texttt{x}` | monospace | | `\verb\|x\|` | verbatim | | `\hrule` | horizontal line | | `\vskip{N}` | N blank lines | | `\hskip{N}` | N spaces | ### Table Syntax ``` \table[width:W, frame:STYLE, pad:{l:L,r:R}, border:STYLE, align:A]{ \row[frame:{b:STYLE}]{ \col[width:N, align:A]{content} } } ``` ### Layout Syntax ``` \begin[width]{vbox} \child[width][h_align]{content} \end{vbox} \begin[top|middle|bottom]{hbox} \child[width][h_align]{content} \end{hbox} ``` ### Settings ``` \setwidth{N} \setunicode{true|false} \setmathitalic{true} \linebreaker{greedy|knuth|raggedright} \sethyphenate{true} \setparskip{N} \setmathabove{N} \setmathbelow{N} ```
    Connector
  • Send a URL and receive the complete HTML of the page after JavaScript execution, exactly as a browser would see it. The request goes through a French residential IP driving a real Chromium instance, which resolves single-page applications and gets past defences that block datacenter ranges. — $0.005/call, paid per request via x402 (USDC).
    Connector
  • Render emailmd markdown into email-safe HTML. Returns html (the complete email document), text (the plain-text MIME part), meta (frontmatter), warnings (non-fatal repairs made while rendering; aim for none), htmlBytes, and previewUrl (a live browser preview of this exact document to share with the user).
    Connector
  • Render a web page to a PNG/JPEG screenshot or a PDF using a real headless browser. Returns the image inline (base64) or the PDF as an embedded resource. Results are cached for 24h.
    Connector
  • Describe a screen, keep the style guide: an LLM turns both into one dense image prompt and an image model renders the mockup. text×2 -> join -> llm -> image; returns image (qwen-image-3) saved to disk (file path in result). Runs on NanoGPT — $0.20 deposit per call, paid in Nano (XNO) — settles at actual model cost + 20%, change returned; no account needed; last run $0.079, ~2m 16s.
    Connector
  • Render a drawing in any supported format (DXF or PDF) to a PNG you can look at, detected from its bytes rather than its name. Use this when you do not know which format you have. Renders the first page/sheet by default; pass `space` (a name from a describe reply's `spaces`) to render another one. Some chat UIs do not display the returned image to the user: for URL sources the result also includes a direct image link — show it (e.g. as a markdown image) when they need to see the render. When the user wants to see or explore the drawing themselves, prefer view_dxf (interactive viewer) — if your platform gates it behind user approval, offer it and ask rather than substituting a static render.
    Connector

Matching MCP Servers

Matching MCP Connectors

  • Screenshot, PDF, and OG-image rendering over MCP. Bearer key or per-call x402 USDC.

  • Fireship-style short videos from a topic — Remotion motion graphics, word-timed captions, 1080p.

  • Render an picsart_media_video_create/picsart_media_video_revise code_url to a final MP4, up to 3840x2160 — use for output above 1920x1920 (the mp_* Scene tools, e.g. picsart_media_export, cap at 1920x1920). Resolutions: hd (1280x720), full_hd (1920x1080, default), ultra_hd (3840x2160, the one output size no mp_* tool can produce). Cost: 5 credits (full_hd), 10 credits (ultra_hd); hd not separately measured, expected no higher than full_hd. IMPORTANT: duration_seconds is currently capped conservatively — 15s for hd/full_hd, 10s for ultra_hd — because a 30s render measured ~32 minutes of real wall time in testing regardless of resolution; requests above the cap are refused up front rather than risking a charged, uncompleted render. If the clip will be composited into an MP Scene afterward (picsart_media_patch_scene, picsart_media_export, etc.), render at hd/full_hd — MP composition itself caps at 1920x1920, so a 4K render only matters as a final, standalone output, never as a scene input.
    Connector
  • Read one clip: its elements (positions/sizes in canvas pixels), voiceover (text, voice, duration, voiceover_volume), background and transition. Pass `render` to also get a PNG of the frame. ASK FOR WHAT YOU NEED. A full read is large — on a dense clip the per-word voiceover array and the element type_data blobs dominate it, and repeated full reads are the main way a long session runs out of context. `select` returns exactly the parts you name: select: ['elements.x','elements.y','elements.width','elements.height'] → geometry only, to fix a layout select: ['elements.name','elements.start_time','elements.end_time'] → a timing pass select: ['words'] → word timings only, to sync visuals to narration select: ['elements.textdata','words'] → rewrite copy against the VO select: ['elements'] → whole element rows, no words select: [] → no JSON at all (pair with render for the PNG alone — smallest read) (omit select) → everything; fine for a first look, expensive to repeat `render` is the other output, and it is separate from `select`: `select` shapes the JSON, `render` produces a PNG. render: {} → the frame at t=0 render: { timestamp: 2.5 } → the frame 2.5s into the clip render: { save: true } → also uploads the PNG and returns presigned_url select: [], render: {} → the PNG alone, no JSON select: ['elements'], render: {} → element rows AND the frame Omitting `render` renders nothing. `timestamp` and `save` live inside it because they only mean anything for a render — there is no way to ask for them without asking for the image. `element_ids` is the other axis: it picks WHICH element rows come back, independently of `select`. Combine them for the leanest read — e.g. element_ids: ['el_9'], select: ['elements.x','elements.y']. Element shape: universal wrapper fields (id, geo, name, x, y, width, height, start_time, end_time, rotation) plus type-specific data (textdata/shapedata/imagedata/videodata/zoomdata) plus an optional `keyframes` array when animated. Keyframes come back in the same flat wire shape add_elements takes — { timestamp, positionX?, positionY?, width?, height?, interpolation? } in canvas pixels — so you can round-trip read → edit → update_elements without reshaping. Clip-level fields include `transition` (the current transition object — sibling of the update_clips `transition` arg; null if none) and `voiceover_words` (per-word timestamps; null on clips with no transcription).
    Connector
  • Save a Hermoso render — or ANY file — into the user’s connected Microsoft OneDrive. Pass a Hermoso render URL as url (or urls[] for several); for a local/external file, call upload_file first and pass the url it returns. Optional folder (created if new) + name. Returns the OneDrive file(s) with a webViewLink. Needs OneDrive connected (Settings ▸ Connectors ▸ OneDrive).
    Connector
  • Download a completed Future Video Studio final render URL to a local file. Use this only after fvs_get_render_status or fvs_get_paid_render_status returns a final_video_url for a completed render. The tool performs an unauthenticated HTTPS GET to that signed URL and writes the response bytes to output_path on the MCP server's local filesystem. It does not call the FVS Agent API, spend wallet credits, require FVS_AGENT_API_KEY, cancel jobs, or modify remote render state. Side effects and constraints: output_path is a local filesystem path for the MCP server process, parent directories are created, existing files are not replaced unless overwrite is true, and large videos may take minutes to download. The request timeout is 600 seconds. Use a fresh status check to refresh expired signed URLs, and do not pass arbitrary or untrusted URLs.
    Connector
  • Render one of a style's two template images — a REAL step of style setup, not an optional extra: a style isn't finished until both its character and environment templates are rendered (the app shows them on the style card). Asset reference images render against them (characters → character template; environments and objects → environment template), and segment renders fall back on them when a shot has no asset reference — so finish BOTH before generate_asset_reference. Run once per template_type ("character" | "environment") for every new style; skip types the style already has (get_style's `templates`). A template already exists is a hard stop here — the call refuses unless replace=True, because overwriting one silently re-anchors every future render. Async — await_jobs(style_id=...), then get_style.
    Connector
  • Render a DXF drawing to a PNG image you can look at. Use this to answer visual questions (what does it look like, where is a feature) — it returns an image, not text. For structural facts and measurements, prefer describe_dxf; never measure pixels. Some chat UIs do not display the returned image to the user: for URL sources the result also includes a direct image link — show it to the user (e.g. as a markdown image) when they need to see the render. When the user wants to see or explore the drawing themselves, prefer view_dxf (interactive viewer) — if your platform gates it behind user approval, offer it and ask rather than substituting a static render.
    Connector
  • Render a PDF drawing's vector content to a PNG you can look at. Images, shadings, and transparency are not drawn — they are reported by describe_pdf — so this shows line work and text, not a page facsimile. For a DXF use render_dxf; if you do not know the format, use render_doc. Renders the first page/sheet by default; pass `space` (a name from a describe reply's `spaces`) to render another one. Some chat UIs do not display the returned image to the user: for URL sources the result also includes a direct image link — show it (e.g. as a markdown image) when they need to see the render. When the user wants to see or explore the drawing themselves, prefer view_dxf (interactive viewer) — if your platform gates it behind user approval, offer it and ask rather than substituting a static render.
    Connector
  • Save a Hermoso render — or ANY file — into the user’s connected Google Drive. Pass a Hermoso render URL as url (or urls[] for several); for a local/external file, call upload_file first and pass the url it returns. Optional folder (created if new) + name. Returns the Drive file(s) with a webViewLink. Needs Google Drive connected (Settings ▸ Connectors ▸ Google Drive — one connection covers Drive, Sheets and Docs). NOTE: Hermoso uses the drive.file scope, so it reaches ONLY the files it created plus any the user explicitly handed over with the Google file picker in the app — never their whole Drive.
    Connector
  • Render a finished ad IMAGE and return its served URL. refImages (local paths or URLs) force product-accurate compositing (drops a real product into the scene). MULTI-BRAND CAUTION: useBrand hydration pulls the SAVED workspace brand — when working a brand that is NOT the saved one (a fresh draft_brand), pass that brand's own productImages/logo as refImages (and useBrand:false) or the output composites the WRONG brand's product. NOTE that the saved-brand hydration also decides the ENGINE: attaching product photos routes the render to the compositing model, so a `model` you named is only honoured when no references ride — pass raw:true (or useBrand:false) to render on exactly the model you asked for. model = a catalog id from hermoso_capabilities (omit for the default). RAW MODEL ACCESS: raw:true dispatches your prompt to the model BYTE-IDENTICAL — no rewriting, no appended guidance, no negative prompt, no brand references attached on your behalf. Credits, the durable delivery of the finished asset and the per-model validation are unchanged. Fast (seconds). Spends credits.
    Connector
  • RECOMMENDED for finished video ADS: render a plan_ad concept through the SAME quality pipeline as the Hermoso web Studio — timed shot list, exact/clean speech (no garbled words), text composited in post (never model-painted), brand end card, licensed music bed, real product references. Pass plan_ad’s full structured output as `creative`. Honors the plan’s render_plan structure/duration: a storyboard that FITS ONE CLIP OF THE RENDER MODEL renders as a single continuous pass; anything longer automatically renders as STITCHED ACTS (the fewest balanced clips, each at most one model clip) — never time-compressed into one clip. That threshold is the render model’s own maximum, not a fixed number: most models cap a clip at 15s and the longest-clip one goes to 30s, so use dryRun:true to see the act split this plan will actually get, for free, before spending. CAST A SAVED CREATOR with `creator` so the SAME person stars in this ad as in the last one (list_creators is the roster) — otherwise every render invents a new face. Renders take 1–3 min; keep polling get_job if it returns still-rendering. Spends credits.
    Connector
  • Render a mingrammer/diagrams Python snippet to PNG and return the image. The code must be a complete Python script using `from diagrams import ...` imports and a `with Diagram(...)` context manager block. Use search_nodes to verify node names and get correct import paths before writing code. Read the diagrams://reference/diagram, diagrams://reference/edge, and diagrams://reference/cluster resources for constructor options and usage examples. Args: code: Full Python code using the diagrams library. filename: Output filename without extension. format: Output format — ``"png"`` (default), ``"svg"``, or ``"pdf"``. download_link: If True, return a temporary download URL path (/images/{token}) that expires after 15 minutes; if False, return inline image bytes. Defaults to True (URL) — set ``DIAGRAMS_INLINE_DEFAULT=true`` on the server to flip the default. SVG/PDF and PNGs larger than the inline limit always use a download link.
    Connector
  • Use this when the user wants to preview or generate a chart. Validate each generated or changed config once before rendering, saving, or updating. Do not revalidate unchanged input; a successful render preflights it for later saves or updates. Errors block; warnings require explicit user approval for the exact config. Suggestions do not block. Replace the complete config with suggestedConfig and revalidate before repair. Stores a 1 hour preview and returns public image URLs. Authenticated previews use the account's render allowance, and each image produced from a returned URL is another render. Anonymous previews are complimentary; their images use the anonymous image allowance.
    Connector
  • Generate a video from images, video clips, or both, synced to an audio track. Use this for narrated question backgrounds, topic visualisations, or any form node that benefits from video. Combine with clipform_generate_tts for narrated audio and clipform_search_media for royalty-free images. Creates 9:16 (720x1280) with Ken Burns pan/zoom effects and transitions. Returns a public URL when complete. Items: type "image" (Ken Burns motion) or "video" (cover-cropped, muted by default). Duration matches audio_url or set duration_seconds explicitly. For multi-question builds, pass wait: false on every render: each call returns a job ID immediately, so all renders run in parallel - then collect URLs with clipform_check_render. Sequential waiting renders take 15-120 seconds EACH. Choosing a render tool: for a recognisable form/quiz beat (guess-the-city, this-or-that, mystery reveal, multiple choice, photo montage...) reach for a video template first (clipform_list_video_templates + clipform_render_video_template) - it is a one-call recipe. Use clipform_generate_video for a narrated or audio-synced media montage (images/clips timed to a voice track). Use clipform_render_composition only when neither fits and you need a custom layer stack. Montage disambiguation: choose clipform_generate_video when the montage is narrated or synced to an audio track; choose the slideshow video template when it is silent (motion + transitions only, no voice-over). A render for a form node is not done until it is attached to that node. Pass node_id (and form_id) so the completed render attaches itself automatically - do not poll clipform_check_render to completion or manually chain clipform_upload_media_asset + clipform_attach_node_media; fire the render and move on.
    Connector
  • Render a specialised video composition to MP4 or PNG - custom animated visuals that clipform_generate_video can't provide, such as geography animations or designed motion graphics. Available composition names and their required input props schemas come from clipform_list_compositions. For narrated Ken Burns slideshows from images, use clipform_generate_video instead. Output formats: mp4 (H.264, best for social media) or png (single frame). Returns a public URL when complete. For multi-render builds (e.g. one clip per quiz question), pass everything in ONE call via items (max 10): all renders fire in parallel and you get one job ID each - then collect the URLs in a single clipform_check_render call with job_ids. Single render: pass compositionId/inputProps at the top level (wait: true blocks and returns the URL; wait: false returns a job ID). Choosing a render tool: for a recognisable form/quiz beat (guess-the-city, this-or-that, mystery reveal, multiple choice, photo montage...) reach for a video template first (clipform_list_video_templates + clipform_render_video_template) - it is a one-call recipe. Use clipform_generate_video for a narrated or audio-synced media montage (images/clips timed to a voice track). Use clipform_render_composition only when neither fits and you need a custom layer stack. A render for a form node is not done until it is attached to that node. Pass node_id (and form_id) so the completed render attaches itself automatically - do not poll clipform_check_render to completion or manually chain clipform_upload_media_asset + clipform_attach_node_media; fire the render and move on. In batch mode, set node_id per item (see items) and form_id once at the top level.
    Connector