Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault

No arguments

Capabilities

Features and capabilities supported by this server

CapabilityDetails
tools
{
  "listChanged": true
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
document

Create, ingest, and inspect documents.

Actions:

  • create: Scaffold a new empty document. Optionally set class, title, author.

  • ingest: Parse markdown text, .tex file, or .md file into the document model.

  • outline: Show document structure (sections, block counts).

  • read: Read content of a specific section as prose text.

  • update: Update document metadata (title, author, date, abstract).

  • reset: Clear the current document and saved state. Next create/ingest starts fresh.

  • bib_add: Add a bibliography entry (provide BibTeX-format entry as source).

  • bib_remove: Remove a bibliography entry by key (provide key as source).

  • bib_list: List all bibliography entries.

  • bib_style: Set bibliography style (provide style name as source, e.g. "authoryear", "numeric").

layout

Configure document typesetting and layout.

Only provided parameters are changed; others are left as-is.

section_break: Page break before level-1 sections. "before" inserts \clearpage before each \section. Works in any column mode. Individual sections can override via edit(page_break="before").

edit

Manipulate document content structurally.

Actions:

  • insert: Add a new block at a position within a section.

  • replace: Replace a block at a position with new content.

  • delete: Remove a block at a position.

  • move: Move a block from one location to another.

  • read_raw: Read a RawLatex block with line numbers.

  • replace_raw: Update a RawLatex block (full or line-level) with lint check.

Sections are addressed by title path (e.g., 'Methods/Data Collection'). Blocks within a section are addressed by 0-based index.

page_break: For sections — "before", "after", "both", or "" to clear. Emits \clearpage before/after the section.

render

Compile and export the document.

Actions:

  • compile: Serialize model to .tex, compile to PDF. Returns PDF path.

  • preview: Render a specific page as PNG file. Returns file path and dimensions.

  • tex: Export the raw .tex source. Returns the LaTeX content.

reference

Search LaTeX documentation, symbols, packages, and error solutions.

Actions:

  • search: Search for LaTeX commands or general topics.

  • symbol: Find symbols by description (e.g., "approximately equal").

  • package: Get information about a LaTeX package.

  • check_style: Analyze a .tex file for best practices.

  • error_help: Get help for LaTeX error messages.

  • example: Get working examples for a topic (table, equation, figure, list, code).

  • templates: Browse available LaTeX templates. Optionally filter by category or slug.

  • capabilities: Enumerate system LaTeX support (engines, tools, packages).

  • styles: Browse document style presets. Optionally filter by slug or keyword.

queue

Execute multiple operations in a single call.

Each operation is a dict with 'tool' (document, layout, edit, render, reference) plus the arguments for that tool. Operations run sequentially; disk is written once at the end.

Example: queue(operations=[ {"tool": "document", "action": "create", "title": "My Doc"}, {"tool": "edit", "action": "insert", "block_type": "section", "title": "Intro", "level": 1}, {"tool": "edit", "action": "insert", "content": "Hello world.", "section": "Intro"}, {"tool": "layout", "columns": 2, "font": "palatino"} ])

Args: operations: List of operation dicts. continue_on_error: If False (default), stop on first error.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription
workflow_mapTeXFlow workflow map showing states and transitions.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/aaronsb/texflow-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server