One MCP — Watchlist | MCP100 Index

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Follow-up value is specific to One action-discovery and execution tools across connected applications: Do compact discovery tools expose the right connected-account authority before executing cross-platform actions?

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

Best fit

Controlled execution of authenticated API actions across connected business applications.

What we know

Integration gateway that discovers/document actions and executes permitted calls across connected applications through local stdio or hosted OAuth MCP.

Maintenance / activity evidence

@withone/mcp 2.0.0 published 2026-09-30T19:42:55.491Z

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.
Potential read/write/admin authority across connected services depending on scopes. Writes: Execute permitted actions in connected apps, including messaging, record changes, commerce/finance and admin operations.
OAuth grant or `ONE_SECRET`; provider tokens are proxied by One.
Sensitive access: Connected-app data, Messages/files/CRM/payment data, Integration grants

Broad integration gateway; risk is driven by connected accounts and method/action scopes.
Default local access should be constrained explicitly.
Use least-privilege consent and connection/action filters.

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

All limitations / Review context

Do compact discovery tools expose the right connected-account authority before executing cross-platform actions?

The bounded review of the publisher's documented One action-discovery and execution tools across connected applications 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.

Broad integration gateway; risk is driven by connected accounts and method/action scopes.
Default local access should be constrained explicitly.
Use least-privilege consent and connection/action filters.
Potential read/write/admin authority across connected services depending on scopes.
OAuth grant or `ONE_SECRET`; provider tokens are proxied by One.
Requests are proxied through One; package docs state platform tokens are not exposed directly to the MCP process and sensitive headers/secrets are redacted from returned config/logging.
Reviewed write actions: Execute permitted actions in connected apps, including messaging, record changes, commerce/finance and admin operations
Sensitive access: Connected-app data, Messages/files/CRM/payment data, Integration grants
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.
retention/deletion details
pricing
Effective capability depends on connected services and grants.
Local mode is broad unless permission filters are explicitly configured.
Catalog/tool counts are live and should not be treated as stable.
Retention/deletion details were not established.

Original activity observations

@withone/mcp 2.0.0 published 2026-09-30T19:42:55.491Z

Technical identity

npm/@withone/mcp

Exact implementation source: https://www.npmjs.com/package/@withone/mcp ↗
Reviewed resource type: Gateway
Repository identity: withoneai/mcp
Package identity: npm:@withone/mcp
Service endpoint: https://mcp.withone.ai/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 ↗
@withone/mcp
Recorded package identity, distribution or download evidence.
Includes references whose attribution is not established.
Evidence source ↗
withone.ai/docs
Source material retained for this review.
Includes references whose attribution is not established.
Evidence source ↗
withone.ai/changelog
Source material retained for this review.
Includes references whose attribution is not established.
GitHub repository ↗
withoneai/mcp
Repository reference; implementation attribution is not established for every document.
Includes references whose attribution is not established.

Setup · Watchlist

Gateway setup

Publisher instructions.

Source evidence observed Oct 14, 2026.

Additional setup details required — see the specific gaps below.

Install / connect gateway
Publisher instructions
Remote OAuth endpoint https://mcp.withone.ai/mcp or local package; One API proxy can act across 700+ connected apps.
Publisher recommends remote MCP https://mcp.withone.ai/mcp with OAuth and scoped consent; local CLI @withone/cli is a different install path.
publisher documented remote registration · Publisher instructions
claude mcp add --transport http one https://mcp.withone.ai/mcp
Configure / authenticate
MCP100 setup guide
Prefer hosted remote OAuth endpoint https://mcp.withone.ai/mcp for One. Local CLI and connection-secret modes are alternatives, not prerequisites for remote OAuth.
remote_streamable_http: https://mcp.withone.ai/mcp
Environment variables referenced by this route (respect the alternatives above): ONE_SECRET
Check connection
Suggested check
Verify OAuth connection and list actions under authorized One connections; do not invoke actions that modify connected apps.
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.
Access grant scopes are selected on One OAuth consent screen; client disconnect not the same as provider-side grant revocation.

Official Documentation & Sources

What it does

Integration gateway that discovers/document actions and executes permitted calls across connected applications through local stdio or hosted OAuth MCP.

Capability: Productivity & Collaboration
Resource type: Gateway
Best fit: Controlled execution of authenticated API actions across connected business applications.

Representative uses

Find and execute actions across connected apps
Use read-only or knowledge-only mode
Scope by identity, connections or action IDs
Generate integration documentation/code

Limitations

Effective capability depends on connected services and grants.
Local mode is broad unless permission filters are explicitly configured.
Catalog/tool counts are live and should not be treated as stable.
Retention/deletion details were not established.

Reviewed access

Access & risk

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

Authentication: OAuth grant or `ONE_SECRET`; provider tokens are proxied by One.
Permissions: Potential read/write/admin authority across connected services depending on scopes.
Data handling: Requests are proxied through One; package docs state platform tokens are not exposed directly to the MCP process and sensitive headers/secrets are redacted from returned config/logging.
Cost: unknown
Risk and cautions: Broad integration gateway; risk is driven by connected accounts and method/action scopes.

Access scope and cautions

Write actions: Execute permitted actions in connected apps, including messaging, record changes, commerce/finance and admin operations
Sensitive access: Connected-app data Messages/files/CRM/payment data Integration grants
Risk level: high
Security notes: Default local access should be constrained explicitly. Use least-privilege consent and connection/action filters.

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: Find and execute actions across connected apps; Use read-only or knowledge-only mode; Scope by identity, connections or action IDs; Generate integration documentation/code.

What setup and authentication are documented?

Recorded installation/entrypoint step: npx @withone/mcp. Recorded endpoint: https://mcp.withone.ai/mcp. Recorded transport: stdio, streamable_http. Recorded environment-variable notes: ONE_SECRET (local); ONE_IDENTITY (optional); ONE_IDENTITY_TYPE (optional); ONE_PERMISSIONS (optional); ONE_CONNECTION_KEYS (optional); ONE_ACTION_IDS (optional); ONE_KNOWLEDGE_AGENT (optional). No configuration values are supplied here. Prerequisites: One account; connected applications; MCP client. Client notes: Use consent scopes or explicit permission/connection/action filters; knowledge-only mode removes execution. These are the documented setup details, not a claim that installation or client compatibility was independently tested. Authentication: Hosted OAuth/consent or local `ONE_SECRET`.

Can it modify data or perform write operations?

Documented write actions include: Execute permitted actions in connected apps, including messaging, record changes, commerce/finance and admin operations. Recorded permission boundary: Potential read/write/admin authority across connected services depending on scopes.

What data-handling boundary should I consider?

Requests are proxied through One; package docs state platform tokens are not exposed directly to the MCP process and sensitive headers/secrets are redacted from returned config/logging. Recorded sensitive-access context: Connected-app data; Messages/files/CRM/payment data; Integration grants.

What limitations and risks should I check before use?

Broad integration gateway; risk is driven by connected accounts and method/action scopes. Effective capability depends on connected services and grants. Local mode is broad unless permission filters are explicitly configured. Catalog/tool counts are live and should not be treated as stable. Retention/deletion details were not established. Default local access should be constrained explicitly. Use least-privilege consent and connection/action filters. 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.

Is this an official MCP?

One / withoneai is listed as First-party MCP in MCP100’s current evidence.