Eve — Watchlist | MCP100 Index

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Follow-up value is specific to Durable agent runtime accepting MCP tools with deployer-configured safeguards: How do preview changes affect MCP approval policies and permissive default execution boundaries?

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

Best fit

Teams building durable backend agents with tools, channels, schedules and sandbox workflows.

What we know

A filesystem-first TypeScript framework for durable agents, tools, channels, schedules, sandboxes and persistent runs.

Maintenance / activity evidence

eve 0.71.1 published 2026-10-05T18:02:03.761Z

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.
Developer-defined tools/channels/sandboxes. Writes: Tools; Hooks; Schedules; Sandbox operations.
Application/channel-defined
Sensitive access: Sessions, Prompts, Workspace files

Durable agents can run scheduled and sandboxed actions; route protection, tool scope and provider configuration determine safety.

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

Popularity

5,474 repository stars · 4.2M package downloads / 30 days

Stars observed Oct 6, 2026. Downloads observed Oct 6, 2026.

All limitations / Review context

How do preview changes affect MCP approval policies and permissive default execution boundaries?

The bounded review of the publisher's documented Durable agent runtime accepting MCP tools with deployer-configured safeguards 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.

Durable agents can run scheduled and sandboxed actions; route protection, tool scope and provider configuration determine safety.
Developer-defined tools/channels/sandboxes.
Application/channel-defined
Durable state and sessions may be persisted; providers receive configured data.
Reviewed write actions: Tools; Hooks; Schedules; Sandbox operations
Sensitive access: Sessions, Prompts, Workspace files
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.
It is a framework, not an MCP server.
Auth, provider credentials and sandbox authority are application-defined.

Original activity observations

eve 0.71.1 published 2026-10-05T18:02:03.761Z

Technical identity

npm/eve

Exact implementation source: https://www.npmjs.com/package/eve ↗
Reviewed resource type: Framework
Repository identity: vercel/eve
Package identity: npm:eve

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 6, 2026

Package registry ↗
eve
Recorded package identity, distribution or download evidence.
Includes references whose attribution is not established.
GitHub repository ↗
vercel/eve
Repository reference; implementation attribution is not established for every document.
Includes references whose attribution is not established.
Evidence source ↗
eve.dev/docs
Source material retained for this review.
Includes references whose attribution is not established.

Setup · Watchlist

Framework setup

Publisher instructions.

Source evidence observed Oct 14, 2026.

Additional setup details required — see the specific gaps below.

Warning

Not runtime tested, no third-party installation, network authorization, paid calls, or mutation executed.
Do not claim production service is available based on README alone.
Agent framework/host with project scaffolding; not a standalone MCP server registration
Targeted 2026-10-13 re-research: Agent framework with npx eve@latest init my-agent; not standalone MCP server.
Agent framework with npx eve@latest init my-agent; not standalone MCP server.
Official npm publisher package page README inspected for this exact npm identity; static research only. Some setup statements are conservatively summarized; no source code or live client execution checked. V2 review did not run any live client.
Avoid write, payment, token transfer, publishing, sending, booking and remote account changes as tests.
Eve agent framework initialized using npx eve@latest init; no evidence of direct MCP stdio binary for this package.
Replace credential, action and registration placeholders; follow the matching publisher/client prerequisites. No execution performed.

Install framework / project integration
Publisher instructions
Agent framework/host with project scaffolding; not a standalone MCP server registration | Publisher npm README revisited on 2026-10-14; exact identity retained.
Reviewed setup example · Publisher instructions
npx eve@latest init my-agent
Configure / authenticate
MCP100 setup guide
Eve is an agent framework; initialize and run its dev workflow, not a standalone MCP stdio server.
Check connection
Suggested check
Verify scaffold/dev workflow; no automatic client MCP registration is implied.
Remove / uninstall / revoke
MCP100 setup guide
Remove the framework/library from the consuming application using its package manager and undo its generated integration configuration. This is NOT a client-registered standalone MCP server.

Official Documentation & Sources

What it does

A filesystem-first TypeScript framework for durable agents, tools, channels, schedules, sandboxes and persistent runs.

Capability: AI & Reasoning
Resource type: Framework
Industry: Software & Technology
Best fit: Teams building durable backend agents with tools, channels, schedules and sandbox workflows.

Representative uses

Scaffold agents
Define typed tools/channels/hooks
Schedule durable workflows
Deploy agent backends

Limitations

It is a framework, not an MCP server.
Auth, provider credentials and sandbox authority are application-defined.

Reviewed access

Access & risk

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

Authentication: Application/channel-defined
Permissions: Developer-defined tools/channels/sandboxes.
Data handling: Durable state and sessions may be persisted; providers receive configured data.
Cost: unknown
Risk and cautions: Durable agents can run scheduled and sandboxed actions; route protection, tool scope and provider configuration determine safety.

Access scope and cautions

Write actions: Tools Hooks Schedules Sandbox operations
Sensitive access: Sessions Prompts Workspace files
Risk level: high

FAQ

Based on reviewed evidence as of 2026-10-06. Watchlist tracking is not a recommendation or a security certification.

Which tasks are documented for this implementation?

The reviewed tasks include: Scaffold agents; Define typed tools/channels/hooks; Schedule durable workflows; Deploy agent backends.

What setup and authentication are documented?

Recorded installation/entrypoint step: npx eve@latest init my-agent. Prerequisites: Node.js/TypeScript; Vercel/Nitro workflows as applicable. These are the documented setup details, not a claim that installation or client compatibility was independently tested. Authentication: Application/channel-defined.

Can it modify data or perform write operations?

Documented write actions include: Tools; Hooks; Schedules; Sandbox operations. Recorded permission boundary: Developer-defined tools/channels/sandboxes.

What data-handling boundary should I consider?

Durable state and sessions may be persisted; providers receive configured data. Recorded sensitive-access context: Sessions; Prompts; Workspace files.

What limitations and risks should I check before use?

Durable agents can run scheduled and sandboxed actions; route protection, tool scope and provider configuration determine safety. It is a framework, not an MCP server. Auth, provider credentials and sandbox authority are application-defined. 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?

Vercel is listed as First-party MCP in MCP100’s current evidence.