Deterministic AST code-intelligence engine for AI agents — MCP server + REST API
Deterministic AST code-intelligence engine for AI agents — MCP server + REST API for architecture analysis, security audits, and doc generation.
active Production-safe Score 88/100 http no auth
At a glance
| Endpoint | https://mrcp-engine.vercel.app/api/mcp |
|---|---|
| Source | https://github.com/faelscarpato/mrcp-engine |
| Transport | http |
| Tools | 27 |
| Auth | none |
| First seen | 2026-08-30 04:16:19 |
| Last checked | 2026-11-17 21:41:43 |
Observed reliability
| 7-day uptime | 100% |
|---|---|
| 24-hour uptime | 100% |
| Handshake success | 100% |
| Latency p50 / p95 | 926 / 1316 ms |
| Last successful check | 2026-11-17 21:41:43 |
Measured by Arclan's validator via repeated live MCP handshakes. How this is computed →
Tools (27)
analyze_repository— [Category: Core Engine] Analyzes the structural AST graph of a GitHub repository. Returns nodes (files, modules, functions), edges (dependenget_repository_skills_contract— [Category: Core Engine] Returns actionable skill contracts for hotspot files in a repository. Each contract includes the detected language,mrcp_run_full_repository_suite— [Category: Core Engine / Full Suite] Runs all 12 repository intelligence tools in a single automated pipeline (AST Graph, Skills Contracts,mrcp_impact_analysis— [Category: Predictive & Security Engineering] Calculates the AST Blast Radius of code changes. Traces downstream impacted files, functions,mrcp_security_compliance_audit— [Category: Predictive & Security Engineering] Performs static AST security audit (OWASP, hardcoded secrets, unsafe execution, package licensmrcp_architectural_drift_detector— [Category: Predictive & Security Engineering] Detects architectural drift, circular dependencies, and cross-layer violations against Clean Amrcp_auto_test_coverage_gap_finder— [Category: Predictive & Security Engineering] Maps high-complexity function nodes against tests to identify uncovered gaps and generate Vitemrcp_context_pruning_pack— [Category: Predictive & Security Engineering] Slices a minimal, hyper-focused AST context prompt pack for a specific task to optimize LLM tomrcp_ast_refactor_applier— [Category: High-Efficiency Agent Offloading] Applies deterministic AST refactoring (rename symbols, extract interfaces, update imports) in bmrcp_type_signature_extractor— [Category: High-Efficiency Agent Offloading] Extracts only TypeScript, Python, Go, Rust, or JS type signatures and interfaces, eliminating imrcp_git_diff_semantic_summarizer— [Category: High-Efficiency Agent Offloading] Strips whitespace/formatting noise from git diffs and summarizes changes by domain and modifiedmrcp_dependency_compatibility_resolver— [Category: High-Efficiency Agent Offloading] Queries the real NPM Registry to resolve package versions, peer dependency conflicts, deprecatimrcp_dead_code_pruner— [Category: High-Efficiency Agent Offloading] Performs AST tree-shaking reachability analysis to identify unreferenced exports, dead variablemrcp_sql_schema_orm_contract_generator— [Category: High-Efficiency Agent Offloading] Parses real Prisma schemas, SQL DDL, or Drizzle/TypeORM models to expose typed DB tables, colummrcp_api_contract_generator— [Category: High-Efficiency Agent Offloading] Scans codebase for API route handlers (Next.js, Express, Fastify, Hono, FastAPI, Flask) and outmrcp_code_metrics_health_scorer— [Category: Predictive & Security Engineering] Calculates Maintainability Index (MI 0-100), technical debt score, cognitive load distributionmrcp_env_secret_contract_validator— [Category: Predictive & Security Engineering] Scans codebase for environment variables (process.env, os.environ), validates against .env.examrcp_monorepo_package_graph_analyzer— [Category: High-Efficiency Agent Offloading] Maps monorepo workspace packages (pnpm, turbo, lerna, nx), internal dependencies, circular linkmrcp_docstring_api_doc_generator— [Category: High-Efficiency Agent Offloading] Generates standardized TSDoc / JSDoc / Python docstrings and markdown API reference tables formrcp_document_analyzer— [Category: Core Engine / Document Intelligence] Parses and analyzes non-code document repositories (CSV, TSV, TXT, MD, DOCX, XLSX, XLS, PDF,mrcp_triage_parse_resume— [Category: Triage & HR] Deterministically extracts candidate name, email, and skills from raw resume text via regex and dictionary parsing.mrcp_triage_score_candidate— [Category: Triage & HR] Calculates candidate match score against a job description using deterministic set theory and keyword overlap.mrcp_triage_generate_hr_report— [Category: Triage & HR] Generates formatted HR triage report summary from AI dossier evaluation.mrcp_web_search— [Category: Web Scraper] Searches DuckDuckGo for relevant links, titles, and snippets without downloading full page bodies.mrcp_web_scrape— [Category: Web Scraper] Fetches a URL and extracts clean sanitized text content, stripping navigation, ads, and scripts.mrcp_web_smart_search— [Category: Web Scraper] Orchestrates search, keyword relevance ranking, and scrapes top N matching pages simultaneously.mrcp_clone_page— [Category: Web & Front-end Intelligence] Clones and reverse-engineers any web page into design tokens, semantic components, layout tree, pur
Validation history
| Checked | Reachable | Handshake | Latency | Score | State |
|---|---|---|---|---|---|
| 2026-11-17 21:41:43 | ✓ | ✓ | 762 ms | 90 | active |
| 2026-11-17 20:56:50 | ✓ | ✓ | 879 ms | 90 | active |
| 2026-11-17 20:27:10 | ✓ | ✓ | 916 ms | 90 | active |
| 2026-11-17 19:42:26 | ✓ | ✓ | 980 ms | 90 | active |
| 2026-11-17 19:03:05 | ✓ | ✓ | 970 ms | 90 | active |
| 2026-11-17 18:41:48 | ✓ | ✓ | 1038 ms | 86 | active |
| 2026-11-17 17:57:20 | ✓ | ✓ | 856 ms | 90 | active |
| 2026-11-17 17:30:12 | ✓ | ✓ | 790 ms | 90 | active |
| 2026-11-17 16:42:16 | ✓ | ✓ | 1003 ms | 86 | active |
| 2026-11-17 15:59:36 | ✓ | ✓ | 784 ms | 90 | active |
| 2026-11-17 15:25:07 | ✓ | ✓ | 893 ms | 90 | active |
| 2026-11-17 14:40:13 | ✓ | ✓ | 799 ms | 90 | active |
| 2026-11-17 13:42:45 | ✓ | ✓ | 924 ms | 90 | active |
| 2026-11-17 13:25:17 | ✓ | ✓ | 970 ms | 90 | active |
| 2026-11-17 12:41:21 | ✓ | ✓ | 703 ms | 90 | active |
Query this programmatically
Agents can pull this record live over the MCP protocol or REST:
get_server("https://mrcp-engine.vercel.app/api/mcp") # via the Arclan MCP endpoint https://arclan.ai/mcp
GET https://arclan.ai/api/registry/mrcp-engine-vercel-app-api-mcp # JSON