New:Introducing Socket Scanning for VS Code Marketplace Extensions.Learn more →
Get Started

void-os

Package Overview
Dependencies
Maintainers
1
Versions
2
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

void-os

void-os

latest
npmnpm
Version
0.1.0
Version published
Weekly downloads
0
Maintainers
1
Weekly downloads
 
Created
Source

void-os

void-os is an operating system for language models and agents in which every datum is a relation in normal form held as MTSV text: a data bank that keeps what a harness sends, tools through which a model reads it and any other file, hooks through which a harness reaches both, and a view through which a person sees the whole of it.

Install

npm install --global void-os

Modules

void-os: the command: the relational model, data bank, tools, hooks, view and system generation wired into the void-os command, each handed what it needs relational model: what a datum is: a relation, written as text and read back, and operated on by the relational algebra format: how a relation is written as text and read back: MTSV, and JSON at void-os's edges parser: how MTSV text becomes sheets generator: how sheets become MTSV text JSON text: how a JSON text becomes the relations of its instances query: which sheets, records and fields a reader names, as the relational algebra's operations, written by position and by filter positions: how positions are written, 2, 1;3, 1-3 or 2-*, counted from 1, and which positions a selection names filter expression: what a filter expression is, and which records it keeps set operations: the traditional set operations: the union of a sheet across files or inputs operations on relations: the less traditional operations on relations: restriction by position or filter, projection by fields, and join on shared columns data bank: how relations are kept and given back: converted in from JSON, inserted, and communicated converter: how the JSON values a schema describes become relations, written as one MTSV file converter › input: what the converter is given: the module specification and the input values, read through JSON text into relations schema: what a module specification supplies, and how its schema is read and applied value: how each input value is read as a JSON text, and when it is rejected transfer: how the input's relations become relations in normal form relation: which relation and column each value is written to key: how each record is identified, and how a place within a value is named order: the order of sheets, records and columns output: how the relations are written as MTSV, at its levels file: which sheets the whole input's one MTSV file holds, and in which order sheet: the name and the header of each sheet record: each instance as one record of its sheet field: how a value is written as a field's text storage: where and how each input is kept: its folder and file, inserted only, each value whole communication: what the data bank gives: names, records by their keys, and what is new tools: the MCP server and the one tool it offers read: how a pathname is read as MTSV: a file, a directory of files, or the data bank read › input: what a pathname names: a file, a directory of files, or the data bank conversion: how a file of another format becomes sheets, and what it leaves behind package: how a package's zip entries and XML parts are read as nodes csv: how a CSV file becomes a sheet json: how a JSON file becomes sheets xlsx: how a workbook becomes sheets ods: how a spreadsheet document becomes sheets sqlite: how a database becomes sheets reply: what every reply states: the report, then the map or the values server: how the tool is offered and called over MCP hooks: how a harness's hook events reach the data bank, and what they give back hook input: how a hook's input is inserted into the data bank hook output: what a hook gives the harness as context, and in which form view: how a person sees the whole data bank: overview first, zoom and filter, then details on demand overview: what the whole data bank holds: each input, its values, and its days zoom: what one input holds: its days, then the turns of a day filter: which records a level keeps, by the filter query gives details: one value in full: every record of every sheet it gives terminal: how a level is laid out on a terminal, and when it is written as MTSV instead system generation: the decisions postponed to release, and how void-os is built, described and installed values: where the Values are read from, and how values.mtsv is written distribution package: how void-os is built, described and installed: a package on PyPI and on npm, and a server in the MCP Registry runner: how a conformance case is run: found by its key, its fixtures placed, its access program called, and what it asserts compared

FAQs

Package last updated on 14 Oct 2026

Related posts