mcp-code-executor-server
An MCP server that exposes a code-execution tool to AI agents, allowing code submissions in multiple languages to be compiled and run in an isolated/sandboxed environment with resource limits and error output.
Score Breakdown
⚙ Agent Friendliness
🔒 Security
The README emphasizes isolated containers, resource limits (CPU/memory/timeouts), and cleanup of temporary files, which are positive security properties for executing untrusted code. However, the provided content does not document transport security requirements (HTTPS/TLS), authentication/authorization, scope/tenant separation, network/file access restrictions inside the sandbox, or secret-handling behavior. Dependency hygiene cannot be verified from the provided material.
⚡ Reliability
Best When
You need an agent-accessible, standardized MCP tool for quick compile-and-run across several languages, and you can deploy/harden the sandbox and resource limits appropriately.
Avoid When
You cannot ensure container isolation, network/file-system restrictions, or you need strong operational guarantees (SLA, observability, documented retry semantics).
Use Cases
- • Run small code snippets from an AI agent (compile-and-run)
- • Provide automated judging/verification for candidate solutions
- • Support multi-language development workflows inside an agent toolchain
- • Debugging assistance by returning compile-time and runtime errors
Not For
- • Running untrusted code without strong sandboxing guarantees and hardened container/host configuration
- • High-throughput or multi-tenant production environments without documented quotas/authn/authz
- • Long-running workloads or persistent stateful execution beyond the request lifecycle
Interface
Authentication
No authentication/authorization mechanism is described in the provided README content. The server is shown running locally and listening on port 8080.
Pricing
Pricing not applicable/unspecified (appears self-hosted).
Agent Metadata
Known Gotchas
- ⚠ The README describes a single 'code-execution' tool and an example request, but does not document the full MCP schema, expected response fields, or tool naming conventions beyond a conceptual description.
- ⚠ Authentication/rate limiting/quotas are not described; agents should assume they may need external throttling and that misuse could overwhelm the sandbox.
- ⚠ Because execution runs untrusted code, agents should be cautious about passing secrets and about requesting operations that may attempt network access or filesystem interaction unless explicitly blocked in the sandbox implementation.
Alternatives
Full Evaluation Report
Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for mcp-code-executor-server.
AI-powered analysis · PDF + markdown · Delivered within 30 minutes
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
Scores are editorial opinions as of 2026-04-04.