hyperweave

mcp
Guvenlik Denetimi
Uyari
Health Uyari
  • License — License: Apache-2.0
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 6 GitHub stars
Code Gecti
  • Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Gecti
  • Permissions — No dangerous permissions requested
Purpose
This tool acts as a headless visual output layer for AI agents. It enables users to generate self-contained SVG graphics via a single API call without requiring JavaScript or external dependencies.

Security Assessment
The overall risk is rated as Low. The automated code scan of 12 files found no dangerous patterns, hardcoded secrets, or requests for dangerous system permissions. The tool does not attempt to execute hidden shell commands. However, it is important to note that the README file calls out to external URLs (hyperweave.app) to fetch images, meaning the tool or its accompanying assets do make network requests to the developer's server.

Quality Assessment
The project is very new and currently has low community visibility with only 6 GitHub stars, meaning it has not been broadly vetted by a large user base. On a positive note, it is actively maintained (the last push was today) and uses the permissive Apache-2.0 license, which makes it legally safe for integration and modification. Developers should expect early-stage software that may undergo frequent changes.

Verdict
Safe to use, though early adoption should proceed with standard caution typical for unvetted, low-visibility tools.
SUMMARY

Headless visual output layer for AI agents. One API call → self-contained SVG. Zero dependencies, works everywhere ⌬

README.md

HyperWeave

Headless visual output layer for AI agents.
One API call → self-contained SVG. No JavaScript. No dependencies. No runtime.
Works everywhere an <img> tag works.

strip


The Problem

When an AI agent needs visual output, it generates React code or HTML that breaks across platforms, carries no brand identity, and is illegible to the next agent in the chain. There's no portable, reliable visual primitive for agents.

HyperWeave is that primitive. Semantic SVGs with embedded CSS state machines, accessibility markup, and machine-readable metadata. The artifact stays live, stays on-brand, and stays legible — whether it's rendered in a GitHub README, Slack, Notion, documentation site, email, VS Code, or terminal. Every surface that renders an <img> tag is a HyperWeave surface.


Genomes — Aesthetic DNA

A genome is a portable, machine-readable aesthetic specification. It encodes the complete visual identity — chromatic system, surface material, motion vocabulary, geometric form language — as a set of CSS custom properties that any agent can consume and apply consistently across every artifact type.

Three production genomes ship today. Custom genome generation via AI skill files coming soon.

automata chrome-horizon brutalist-emerald
Signals pypi teal pypi amethyst
/v1/badge/PYPI/v0.2.11/automata.static?family={blue|purple}
passing warning critical
/v1/badge/BUILD/{state}/chrome-horizon.static
passing warning critical
/v1/badge/BUILD/{state}/brutalist-emerald.static
Dashboard strip
/v1/strip/readme-ai/automata.static?family=bifamily
strip
/v1/strip/readme-ai/chrome-horizon.static
strip
/v1/strip/readme-ai/brutalist-emerald.static
Profile cards stats
/v1/stats/eli64s/automata.static
stats
/v1/stats/eli64s/chrome-horizon.static
stats
/v1/stats/eli64s/brutalist-emerald.static
star chart
/v1/chart/stars/eli64s/readme-ai/automata.static
star chart
/v1/chart/stars/eli64s/readme-ai/chrome-horizon.static
star chart
/v1/chart/stars/eli64s/readme-ai/brutalist-emerald.static
Marquee marquee horizontal
/v1/marquee/{title}/automata.static?rows=1&family=bifamily
marquee horizontal
/v1/marquee/{title}/chrome-horizon.static?rows=1
marquee horizontal
/v1/marquee/{title}/brutalist-emerald.static?rows=1
marquee vertical
…/automata.static?rows=1&direction=up&family=bifamily
marquee vertical
…/chrome-horizon.static?rows=1&direction=up
marquee vertical
…/brutalist-emerald.static?rows=1&direction=up
Icons python huggingface docker jupyter
/v1/icon/{glyph}/automata.static?shape={circle|square}&family={blue|purple}
youtube notion npm instagram
/v1/icon/{glyph}/chrome-horizon.static?shape={circle|square}
discord github x spotify
/v1/icon/{glyph}/brutalist-emerald.static?shape={circle|square}
Banner banner
/v1/banner/HYPERWEAVE/automata.static?family=bifamily
banner
/v1/banner/HYPERWEAVE/chrome-horizon.static
banner
/v1/banner/HYPERWEAVE/brutalist-emerald.static
automata chrome-horizon brutalist-emerald
Surface #04060A void #000a14 deep void #14532D dark field
Signal #1E849A teal · #6B3B8A amethyst #5ba3d4 metallic blue #10B981 emerald
Profile brutalist (cellular paradigm) chrome (smooth, env-mapped) brutalist (sharp, zero-radius)
Chromatic variant blue / purple / bifamily
Motions static, chromatic-pulse, rimrun, corner-trace 5 border SMIL 5 border + 9 kinetic

The two automata signal badges share a URL and differ only in the chromatic variant: ?family=blue renders the teal palette, ?family=purple the amethyst, ?family=bifamily both. The parameter name is provisional and likely to be renamed to ?variant= once brutalist skin variants land — treat the surface as unstable.

Error fallback — SMPTE NO SIGNAL

Every broken <img> URL renders the SMPTE RP 219 test pattern with ERR_NNN matching the HTTP status, instead of a browser broken-image icon.

404 error fallback (intentionally broken URL)

/v1/badge/TEST/value/nonexistent-genome.static — unknown genome → HTTP 404, value slab reads ERR_404.


Install

uv add hyperweave
# or
pip install hyperweave

Requires Python 3.12+.


Entry Points

Four interfaces, one pipeline. Every path produces the same artifact through the same compositor.

MCP CLI
HTTP API Python SDK

MCP

{
  "mcpServers": {
    "hyperweave": {
      "command": "hyperweave",
      "args": ["mcp"]
    }
  }
}
hw_compose(type="badge", title="build", value="passing", genome="brutalist-emerald")
hw_live(provider="github", identifier="anthropics/claude-code", metric="stars")
hw_kit(type="readme", genome="brutalist-emerald", badges="build:passing")
hw_discover(what="all")

CLI

# Badge
hyperweave compose badge "build" "passing" --genome brutalist-emerald

# Strip with metrics
hyperweave compose strip "readme-ai" "STARS:2.9k,FORKS:278" -g brutalist-emerald

# Banner with kinetic motion
hyperweave compose banner "HYPERWEAVE" -g brutalist-emerald -m cascade

# Artifact kit
hyperweave kit readme -g brutalist-emerald --badges "build:passing,version:v0.2.0" --social "github,discord"

# Profile card (live GitHub data)
hyperweave compose stats eli64s -g chrome-horizon -o stats.svg

# Star history chart
hyperweave compose chart stars eli64s/readme-ai -g brutalist-emerald -o chart.svg

# Timeline / roadmap from JSON items
hyperweave compose timeline --data roadmap.json -g chrome-horizon -o timeline.svg

# Custom genome from a local JSON file (validated against the profile contract)
hyperweave compose badge "DEPLOY" "live" --genome-file ./my-genome.json
hyperweave validate-genome ./my-genome.json

HTTP API

# URL grammar: /v1/{type}/{title}/{value}/{genome}.{motion}
curl https://hyperweave.app/v1/strip/readme-ai/brutalist-emerald.static?value=STARS:2.9k,FORKS:278

# Live data binding
curl https://hyperweave.app/v1/live/github/anthropics/claude-code/stars/chrome-horizon

# POST compose
curl -X POST https://hyperweave.app/v1/compose \
  -H "Content-Type: application/json" \
  -d '{"type":"banner","title":"HYPERWEAVE","genome":"brutalist-emerald","motion":"drop"}'

# Local server
hyperweave serve --port 8000

Session Telemetry

HyperWeave parses Claude Code transcripts into visual receipts — cost, tokens, tool distribution, cognitive phases. The artifact isn't a visualization of data. It is the record.

# Manual
hyperweave session receipt .claude/session.jsonl -o receipt.svg

# Autonomous — install once, every session gets a receipt
hyperweave install-hook

After install-hook, every Claude Code session automatically drops a receipt SVG into .hyperweave/receipts/. No config, no server, no manual step.

session receipt

rhythm strip


How It Works

Every artifact is the output of a single composition formula:

ARTIFACT = FRAME × PROFILE × GENOME × SLOTS × MOTION × ENVIRONMENT

Python builds context dicts. Jinja2 builds SVG. YAML defines config. Three layers, no mixing. Zero f-string SVG in Python.

ComposeSpec → engine.py → assembler.py (CSS) → lanes.py (validate) → templates.py (Jinja2) → SVG

Every artifact ships with:

  • Semantic metadata — provenance, reasoning, spatial trace, aesthetic DNA. Machine-readable context so the next agent in the chain knows what it's looking at and why.
  • CSS state machinesdata-hw-status, data-hw-state, data-hw-regime drive visual transitions through the Custom Property Bridge. No JavaScript.
  • Pure CSS/SMIL animation — all motion uses compositor-safe properties (transform, opacity, filter). No script tags. Works inside GitHub's Camo proxy, email clients, Notion embeds — anywhere SVGs render.
  • Accessibility — WCAG AA, prefers-reduced-motion, prefers-color-scheme, forced-colors, ARIA markup. Structural, not decorative.
Dimension Count
Frame types 15 (badge, strip, banner, icon, divider, marquee-h/v/counter, receipt, rhythm-strip, master-card, catalog, stats, chart, timeline)
Genomes 3 (automata, brutalist-emerald, chrome-horizon)
Motion configs 15 (1 static + 5 border SMIL + 9 kinetic CSS)
Glyphs 99 (93 Simple Icons + 6 geometric)
Divider variants 6 (block, cellular-dissolve, current, takeoff, void, zeropoint)
Metadata tiers 5 (Tier 0 silent → Tier 4 reasoning)
Paradigms 4 per frame (default, brutalist, chrome, cellular) — per-frame dispatch from genome
Bundled fonts 3 (JetBrains Mono, Orbitron, Chakra Petch — base64-embedded)

Stack: Pydantic, FastAPI, FastMCP v3, Jinja2, Typer.


Roadmap

roadmap


Contributing

HyperWeave is early. If you're interested in building genomes, extending frame types, or just seeing what this looks like in your own README — join the Discord.


InnerAura Labs

Discord   Instagram   LinkedIn   TikTok   X   YouTube

Yorumlar (0)

Sonuc bulunamadi