DebugMCP

DebugMCP is a VS Code extension that runs a local MCP server to control the VS Code debugger via AI tools. It lets agents start/stop debug sessions, manage breakpoints, step/continue/restart execution, and inspect variables/evaluate expressions during debugging across multiple languages (via the corresponding VS Code debug extensions).

Evaluated Mar 30, 2026 (0d ago)

Repo ↗ DevTools mcp vscode debugging ai-agents breakpoints step-through variable-inspection expression-evaluation typescript local-first

⚙ Agent Friendliness

/ 100

Can an agent use this?

🔒 Security

/ 100

Is it safe for agents?

⚡ Reliability

/ 100

Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality

Documentation

Error Messages

Auth Simplicity

Rate Limits

🔒 Security

TLS Enforcement

Auth Strength

Scope Granularity

Dep. Hygiene

Secret Handling

Local-first MCP server with no credentials required is claimed; authentication/authorization and scope controls are not described. README mentions secure communication, but details (e.g., transport security for local HTTP vs streamable HTTP) are not specified. Uses zod for validation (suggesting input validation).

⚡ Reliability

Uptime/SLA

Version Stability

Breaking Changes

Error Recovery

Best When

You want an MCP-capable AI assistant to drive interactive debugging in a local VS Code environment with the appropriate language debug extensions installed.

Avoid When

You need strong, user-governed access control for debugger control across machines or you cannot run VS Code locally.

Use Cases

• AI-assisted debugging workflows inside VS Code (set breakpoints, step through code, inspect variables)
• Root-cause analysis of failing tests by starting debug sessions at specific test names
• Multi-language debugging control for agent-based coding assistants that support MCP

Not For

• Cloud-only environments where VS Code debugging cannot run locally
• Use cases requiring hardened authentication/authorization for remote clients (the server is local by design)
• Scenarios needing a hosted API endpoint with documented SLAs

Interface

REST API

GraphQL

gRPC

MCP Server

Yes ↗

SDK

Yes

Webhooks

Authentication

OAuth: No Scopes: No

The README states it runs 100% locally with no credentials required; authentication/authorization for MCP calls is not described.

Pricing

Free tier: No

Requires CC: No

No pricing information provided; distributed as a VS Code extension.

Agent Metadata

Pagination

none

Idempotent

False

Retry Guidance

Not documented

Known Gotchas

⚠ Tools are state-dependent (e.g., stepping/inspecting likely requires an active debug session).
⚠ Agents may need auto-approval/configuration in the client to avoid stalled tool calls.
⚠ Port conflicts require changing serverPort and updating the assistant’s MCP URL.

Alternatives

Using native VS Code debugging with an agent that can edit launch.json and reasoning-only guidance (no debugger control) Manual debugging (human-in-the-loop) with LLM-generated hypotheses and developer-led breakpoint/inspection Other MCP tools that provide code inspection but not direct debugger control

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for DebugMCP.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

API endpoint ↗ Agent guide ↗ Report inaccuracy

Scores are editorial opinions as of 2026-03-30.