UTCP Code Mode — Watchlist

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Track protocol and execution-boundary changes as code-mode evolves

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

Best fit

Developers building agents that need efficient multi-tool workflows and progressive tool discovery.

What we know

A TypeScript framework that lets agents execute code against registered UTCP, MCP, HTTP, file and CLI tools through a compact code-mode surface.

Maintenance / activity evidence

Observed Aug 25, 2026 (UTC)Merge pull request 47 from universal-tool-calling-protocol/dev

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

Activity source · commits ↗Activity source · releases ↗

Verification gaps / watch-outs

Installation and comprehensive security have not been independently audited.
Registered-tool scope. Writes: Depends on registered tools.
Downstream/manual-specific
Sensitive access: Inherited from registered tools

Code mode can orchestrate powerful downstream tools; sandboxing helps but does not replace downstream least privilege.

Security source · repository notice ↗

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

Popularity

1,592 repository stars · 5k package downloads / 30 days

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

All limitations / Review context

How does the current framework bind MCP tools and constrain code execution?

The bounded assessment reviewed the official README, repository notices, security policy path, and security page for known material public-tracking blockers. No such blocker was identified. This is a public-tracking assessment, not an installation safety certification.

Code mode can orchestrate powerful downstream tools; sandboxing helps but does not replace downstream least privilege.
Registered-tool scope.
Downstream/manual-specific
Tool data is processed in sandboxed code execution.
Reviewed write actions: Depends on registered tools
Sensitive access: Inherited from registered tools
Installation, client compatibility and comprehensive security have not been independently audited. Authentication and access depend on the configured services.
A project-specific SECURITY.md policy was not available at the fetched official path; the review remains limited to published repository notices.
No installation, runtime, client matrix, or complete tool-contract audit was performed in this review.
The fetched official SECURITY.md path returned HTTP 404; it was not treated as positive evidence.
Downstream tool permissions and credentials determine effective authority.
Sandbox guarantees are limited to the framework's documented implementation.

Original activity observations

2026-08-25T16:39:22Z: Merge pull request 47 from universal-tool-calling-protocol/dev

Technical identity

github.com/universal-tool-calling-protocol/code-mode

Exact implementation source: https://github.com/universal-tool-calling-protocol/code-mode ↗
Reviewed resource type: Framework
Repository identity: universal-tool-calling-protocol/code-mode
Package identity: npm:@utcp/code-mode

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

GitHub repository ↗
universal-tool-calling-protocol/code-mode
Recorded repository material and project history.
Package registry ↗
@utcp/code-mode
Recorded package identity, distribution or download evidence.
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.

Install / connect
Publisher instructions
UTCP Code Mode is a TypeScript tool-calling library and CLI.
package install · Publisher instructions
npm install @utcp/code-mode
Configure / authenticate
MCP100 setup guide
SDK for Node apps; publisher recommends separate `utcp` CLI for shell-capable agents. MCP-only clients should use its *separate* code-mode-mcp server documented upstream.
Check connection
Suggested check
Verify `@utcp/code-mode` integration via a project example, not by invoking a nonexistent server binary.
Remove / uninstall / revoke
MCP100 setup guide
Remove `@utcp/code-mode` from the Node project with `npm uninstall @utcp/code-mode` and delete its application imports. If you installed the separate Code Mode MCP server in Claude Desktop, remove that **server's** registration separately. The primary CLI path does not need MCP unregistration.
Client unregistration, program uninstall, persistent-data deletion, and credential revocation must be treated separately.
Package / application uninstall · matching installation environment · MCP100 composed guide
npm uninstall @utcp/code-mode

Official Documentation & Sources

What it does

A TypeScript framework that lets agents execute code against registered UTCP, MCP, HTTP, file and CLI tools through a compact code-mode surface.

Capability: AI & Reasoning
Resource type: Framework
Industry: Software & Technology
Best fit: Developers building agents that need efficient multi-tool workflows and progressive tool discovery.

Representative uses

Chain multiple tool calls in one execution
Search tools on demand
Load generated TypeScript interfaces
Process large tool results before returning output

Limitations

Downstream tool permissions and credentials determine effective authority.
Sandbox guarantees are limited to the framework's documented implementation.

Reviewed access

Access & risk

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

Authentication: Downstream/manual-specific
Permissions: Registered-tool scope.
Data handling: Tool data is processed in sandboxed code execution.
Cost: free
Risk and cautions: Code mode can orchestrate powerful downstream tools; sandboxing helps but does not replace downstream least privilege.

Access scope and cautions

Write actions: Depends on registered tools
Sensitive access: Inherited from registered tools
Risk level: moderate

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: Chain multiple tool calls in one execution; Search tools on demand; Load generated TypeScript interfaces; Process large tool results before returning output.

What authentication is documented?

Delegated to registered tools and servers.

Can it modify data or perform write operations?

Documented write actions include: Depends on registered tools. Recorded permission boundary: Registered-tool scope.

What data-handling boundary should I consider?

Tool data is processed in sandboxed code execution. Recorded sensitive-access context: Inherited from registered tools.

What limitations and risks should I check before use?

Code mode can orchestrate powerful downstream tools; sandboxing helps but does not replace downstream least privilege. Downstream tool permissions and credentials determine effective authority. Sandbox guarantees are limited to the framework's documented implementation. A project-specific SECURITY.md policy was not available at the fetched official path; the review remains limited to published repository notices.

Is this an official MCP?

Universal Tool Calling Protocol is listed as First-party MCP in MCP100’s current evidence.