Voxell Forge MCP — Watchlist

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Follow-up value is specific to Forge embedding and model-list tools: Do model dimensions and usage metadata remain documented separately from search, storage and RAG products?

Reviewed Oct 5, 2026 · Documentation review; complete setup workflow not independently tested

Best fit

Hosted text embeddings for RAG, semantic search, clustering and similarity workflows.

What we know

Small MCP around Voxell Forge's hosted embedding API. It exposes embedding generation and model discovery only; it does not provide vector storage/search/RAG itself. Official docs say request text/vectors are not stored, while usage metadata is retained for billing.

Maintenance / activity evidence

@voxell/forge-mcp 0.1.6 published 2026-09-27T02:34:11.652Z

Tracking evidence observed Oct 5, 2026. These dated signals do not establish current maintenance. Reviewed enrichment confidence: high.

Verification gaps / watch-outs

Installation and comprehensive security have not been independently audited.
Embedding-only API calls.
Forge API key.
Sensitive access: Text being embedded may contain sensitive documents, API key

No write authority beyond billing usage, but embedding confidential text sends it to a hosted service.

Full authority, access and evidence limits are available in Review context and Access & risk.

All limitations / Review context

Do model dimensions and usage metadata remain documented separately from search, storage and RAG products?

The bounded review of the publisher's documented Forge embedding and model-list tools and release notices identified no disclosed material condition preventing an accurately scoped public tracking entry. The resource's operational permissions remain uncertainty; this assessment does not certify installation safety.

No write authority beyond billing usage, but embedding confidential text sends it to a hosted service.
Embedding-only API calls.
Forge API key.
Text is sent to Forge; official docs say request data/vectors are not stored, while token/usage metadata is retained.
Sensitive access: Text being embedded may contain sensitive documents, API key
Installation, client compatibility and comprehensive security have not been independently audited. Authentication and access depend on the configured services.
Runtime behavior, dependencies and undisclosed vulnerabilities were not independently audited.
Documented tool authority and any remote data handling must be checked before installation or use.
Tool inventory, client matrix and permissions completeness were not requirements for this Radar tracking review.
Requires Forge API key.
Embeddings are remote API calls; no local inference.
No storage/search layer is included.
Usage cost/quality depends on selected model/tier.

Original activity observations

@voxell/forge-mcp 0.1.6 published 2026-09-27T02:34:11.652Z

Technical identity

npm/@voxell/forge-mcp

Exact implementation source: https://www.npmjs.com/package/@voxell/forge-mcp ↗
Reviewed resource type: MCP Server
Repository identity: VoxellInc/forge-mcp
Package identity: npm:@voxell/forge-mcp

Evidence & Sources

Evidence basis: Recorded source evidence
Setup runtime verification: Not independently runtime-tested by MCP100 for this setup workflow.

Reviewed Oct 5, 2026 · Latest source observation Oct 8, 2026

Package registry metadata ↗
@voxell/forge-mcp
Recorded package identity, distribution or download evidence.
GitHub repository ↗
voxellinc/forge-mcp
Repository reference; implementation attribution is not established for every document.
Includes references whose attribution is not established.
Evidence source ↗
voxell.ai/forge/recipes/
Source material retained for this review.
Includes references whose attribution is not established.

Setup · Watchlist

Install & Agents

Publisher instructions.

Source evidence observed Oct 14, 2026.

Additional setup details required — see the specific gaps below.

Install / connect
Publisher instructions
Replace placeholder and pass key through client env.
publisher documented install or connect · Publisher instructions
claude mcp add forge -e FORGE_API_KEY=<your-key> -- npx -y @voxell/forge-mcp
Advanced configuration · publisher documented step or launch · Publisher instructions
publisher documented step or launch · Publisher instructions · Review before use
claude mcp add forge -e FORGE_API_KEY=YOUR_KEY -- npx -y @voxell/forge-mcp
Configure / authenticate
MCP100 setup guide
Hosted embeddings may consume service usage; list tools or account/status endpoints rather than submitting production content solely for smoke test.
Environment variables referenced by this route (respect the alternatives above): FORGE_API_KEY
Check connection
Suggested check
User-side suggested check (not executed): inspect registered tools/connection and perform publisher-supported read-only status/list operations. Do not send, purchase, publish, delete, sign, trade or alter data as a smoke test.
Remove / uninstall / revoke
MCP100 setup guide
Generic client guidance (not publisher-specific uninstall): unregister MCP entry in the chosen client. If installed globally, remove the npm package separately. Revoke provider access separately through account controls, where supported. This does not erase local data or revoke funds/credentials automatically.
Unregister the named MCP entry from the actual host separately from removing a global npm install; no claim that either revokes remote credentials.
Remove local stored keys/tokens and revoke or rotate corresponding provider authorization using documented provider account controls if available; exact procedure not verified for this package. Wallet private keys cannot be revoked.

Official Documentation & Sources

What it does

Small MCP around Voxell Forge's hosted embedding API. It exposes embedding generation and model discovery only; it does not provide vector storage/search/RAG itself. Official docs say request text/vectors are not stored, while usage metadata is retained for billing.

Capability: AI & Reasoning
Resource type: MCP Server
Best fit: Hosted text embeddings for RAG, semantic search, clustering and similarity workflows.

Representative uses

Embed documents/queries
List available embedding models/dimensions
Build downstream semantic search or RAG with an external vector store
Generate Matryoshka-truncated vectors

Limitations

Requires Forge API key.
Embeddings are remote API calls; no local inference.
No storage/search layer is included.
Usage cost/quality depends on selected model/tier.

Reviewed access

Access & risk

Review the exact access boundary before connecting this MCP to accounts, credentials, or sensitive data.

Authentication: Forge API key.
Permissions: Embedding-only API calls.
Data handling: Text is sent to Forge; official docs say request data/vectors are not stored, while token/usage metadata is retained.
Cost: freemium
Risk and cautions: No write authority beyond billing usage, but embedding confidential text sends it to a hosted service.

Access scope and cautions

Write actions: None in the reviewed inventory
Sensitive access: Text being embedded may contain sensitive documents API key
Risk level: moderate

FAQ

Based on reviewed evidence as of 2026-10-08. Watchlist tracking is not a recommendation or a security certification.

Which tasks are documented for this implementation?

The reviewed tasks include: Embed documents/queries; List available embedding models/dimensions; Build downstream semantic search or RAG with an external vector store; Generate Matryoshka-truncated vectors.

What setup and authentication are documented?

Recorded installation/entrypoint step: npx -y @voxell/forge-mcp. Recorded transport: stdio. Recorded environment-variable notes: FORGE_API_KEY. No configuration values are supplied here. Prerequisites: Node.js >=18; Voxell Forge account/key. These are the documented setup details, not a claim that installation or client compatibility was independently tested. Authentication: Forge API key.

Can it modify data or perform write operations?

The reviewed inventory lists no write actions. An empty inventory alone does not establish read-only authority. Recorded permission boundary: Embedding-only API calls.

What data-handling boundary should I consider?

Text is sent to Forge; official docs say request data/vectors are not stored, while token/usage metadata is retained. Recorded sensitive-access context: Text being embedded may contain sensitive documents; API key.

What limitations and risks should I check before use?

No write authority beyond billing usage, but embedding confidential text sends it to a hosted service. Requires Forge API key. Embeddings are remote API calls; no local inference. No storage/search layer is included. Usage cost/quality depends on selected model/tier. Runtime behavior, dependencies and undisclosed vulnerabilities were not independently audited. Documented tool authority and any remote data handling must be checked before installation or use.