vestige/crates/vestige-mcp
Sam Valladares cfe8d03d36 feat(landing): full-viewport GPU node-engine hero + waitlist + dashboard auth removal
The /launch landing page (bleeding-edge waitlist hero):
- Full-viewport WebGL node engine (src/lib/hero/nodeEngine.ts): 40k GPU particles
  on a two-FBO GPUComputationRenderer running an 18s looping cinematic — particles
  stream in from the screen edges, slam together and EXPLODE at center, reform into
  a brain / graph constellation / neural lattice, dissolve back out, loop. Glowing
  additive particles + UnrealBloom, fills the whole screen edge to edge.
- Key GPGPU fix: custom DataTexture shape targets read black in GPUComputationRenderer
  (three.js #15882), so shape targets are computed PROCEDURALLY in-shader from the
  per-particle seed. Position integrates raw per-frame velocity (Codrops pattern),
  texel-center reference attribute.
- AmbientField (god-ray glow + parallax starfield), phantomBrain (deterministic
  seed-from-identity memory graph), LandingHero/neuralFlow (WebGPU Cinema-storm
  reuse + WebGL fallback, kept as alternates). Manifesto copy, seed input, email
  capture. No em-dashes. SSR off + prerender for the WebGL route.

Waitlist backend (Supabase, owned data):
- supabase/migrations/0001_waitlist.sql (RLS-locked table, dedup), Edge Function
  waitlist-join (CORS, validation, honeypot, dedup, Resend confirmation), setup doc.

Dashboard auth removed (reverts the launch-polish auth wall that 401'd the dashboard
against itself): mod.rs/state.rs/websocket.rs back to no-token, api.ts/websocket.ts/
+layout.svelte stripped of the token client. Pending-review UX + Memory PR gating kept.

Research/spec docs under docs/launch/. Frontend typechecks 0/0.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-24 22:54:51 -05:00
..
src feat(landing): full-viewport GPU node-engine hero + waitlist + dashboard auth removal 2026-06-24 22:54:51 -05:00
Cargo.toml feat(cloud-sync): HTTP managed-sync backend + vestige sync --cloud 2026-06-24 16:13:30 -05:00
README.md Add ComposedGraph composition ledger 2026-06-18 16:00:29 -05:00

Vestige MCP Server

Local cognitive memory for MCP-compatible AI agents.

This crate provides the vestige-mcp stdio MCP server plus the vestige CLI. The cognitive engine lives in vestige-core; this crate owns protocol handling, tool dispatch, optional dashboard serving, backups, restore, update, and portable import/export commands.

Install

For normal users, prefer the release package:

npm install -g vestige-mcp-server

For local development:

cargo build --release -p vestige-mcp

Register With An MCP Client

Use the command vestige-mcp in any stdio MCP client:

{
  "mcpServers": {
    "vestige": {
      "command": "vestige-mcp"
    }
  }
}

Examples:

claude mcp add vestige vestige-mcp -s user
codex mcp add vestige -- vestige-mcp

Transports

  • Default: JSON-RPC 2.0 over stdio.
  • Optional: MCP-over-HTTP on /mcp, enabled only with --http, --http-port, or VESTIGE_HTTP_ENABLED=1.
  • Dashboard: vestige dashboard or VESTIGE_DASHBOARD_ENABLED=1.

HTTP and dashboard bearer tokens are generated locally; see docs/CONFIGURATION.md.

Current Tool Surface

The server exposes the current unified MCP tools from src/server.rs, including:

  • session_context
  • search, smart_ingest, memory, codebase, intention
  • deep_reference, cross_reference, contradictions
  • dream, explore_connections, predict
  • memory_health, memory_graph, composed_graph, system_status
  • importance_score, find_duplicates
  • consolidate, memory_timeline, memory_changelog
  • backup, export, restore, gc, suppress

See the root README.md and docs/AGENT-MEMORY-PROTOCOL.md for agent instructions.

License

AGPL-3.0-only