{"id":"aplavin-julia-mcp","name":"julia-mcp","homepage":null,"repo_url":"https://github.com/aplavin/julia-mcp","category":"ai-ml","subcategories":[],"tags":["mcp","julia","repl","python","stdio","persistent-sessions","agent-tools"],"what_it_does":"Provides an MCP server (stdio transport) that runs Julia code in persistent Julia REPL sessions, keeping state (variables/functions/loaded packages) across tool calls and isolating sessions per environment/project directory.","use_cases":["Interactive code execution for AI assistants with fast iteration (reduced Julia startup/compilation overhead)","Running tests or exploratory analysis in an isolated Julia environment per project directory","Maintaining long-lived Julia state across multiple agent steps"],"not_for":["Untrusted multi-tenant environments where arbitrary Julia execution is unsafe","Production SaaS deployment without additional sandboxing/controls","Use cases requiring strict auditability or limited capability execution"],"best_when":"You have a trusted developer workflow and want persistent Julia sessions to speed up iterative agent-driven computation.","avoid_when":"You need to safely execute code from unknown/untrusted users or require strong isolation between executions/tenants.","alternatives":["MCPRepl.jl","REPLicant.jl","DaemonConductor.jl"],"af_score":63.8,"security_score":39.5,"reliability_score":27.5,"package_type":"mcp_server","discovery_source":["github"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T15:23:34.115966+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":null,"has_sdk":false,"sdk_languages":[],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["None described (stdio-based local MCP server)"],"oauth":false,"scopes":false,"notes":"No authentication mechanism is described; access is presumably controlled by network/process permissions of the client/server setup."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"Open-source tool; runtime costs depend on your infrastructure and Julia workloads."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":63.8,"security_score":39.5,"reliability_score":27.5,"mcp_server_quality":88.0,"documentation_accuracy":78.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":95.0,"rate_limit_clarity":20.0,"tls_enforcement":95.0,"auth_strength":5.0,"scope_granularity":10.0,"dependency_hygiene":35.0,"secret_handling":60.0,"security_notes":"Security posture is largely based on local/stdio transport (no network ports implied) and lack of described authz/auth. Because the tool executes arbitrary Julia code, strong sandboxing/capability restriction is important for any untrusted inputs. Dependency hygiene and explicit hardening details are not provided in the supplied materials.","uptime_documented":0.0,"version_stability":25.0,"breaking_changes_history":20.0,"error_recovery":65.0,"idempotency_support":"false","idempotency_notes":"No explicit idempotency guarantees; repeated calls may mutate persistent session state.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Persistent state can cause surprising effects if an agent assumes a clean environment; use julia_restart when needed.","Arbitrary Julia execution implies the agent must be constrained (policy/tooling) to avoid unsafe code.","Timeout behavior is only partially described (timeout default 60s; auto-disabled for Pkg operations); long-running code may still require careful handling."]}}