Moltline Code Review logo

Moltline Code Review

by GARPHENGATEOfficialWebsiteUpdated Sep 5, 2026

Moltline Code Review & AI-Code Verifier over MCP: review diffs and files the way a skeptical senior engineer would. Free: review_diff risk-scans a unified diff, ai_code_smell_scan flags AI-generated-code tells, plus complexity_report, secret_scan and review_checklist. Premium: security_deep_dive and get_reviewer_persona, unlocked by a Moltline All-Access licence from moltlinestudio.com. No external APIs - paste a diff and get findings. Connects directly to our own endpoint, no signup.

code-review
ai-code
security
+2
|

CONNECT • Endpoint: https://mcp.moltlinestudio.com/codereview (MCP Streamable HTTP) • Claude Code: claude mcp add --transport http moltline-codereview https://mcp.moltlinestudio.com/codereview • Cursor, Codex CLI and any other Streamable HTTP client take the same URL unchanged • Free tools need no account and no API key; every tool description says FREE. or PREMIUM (...) before you call it • Machine-readable server card: https://mcp.moltlinestudio.com/codereview/.well-known/mcp/server-card.json

OVERVIEW A skeptical senior engineer's toolkit for your agent: pattern analyzers that cite the exact line, say why it matters, and give an honest verdict. No external APIs — paste a diff or a file and get findings. Pattern matching only: a clean result is not proof, and every hit deserves human eyes.

KEY CAPABILITIES • review_diff — risk-scans a unified diff for injection sinks, disabled TLS, bare excepts, debug prints, TODOs, N+1 hints and leaked secrets → line-cited findings with severity 1-4, a 0-100 risk score and a verdict • ai_code_smell_scan — flags the tells of unreviewed AI output (comments that restate the next line, leaked assistant preambles, placeholder TODOs, shipped 'Example usage' blocks, error-swallowing try/except) → a reviewed-confidence score with quoted evidence • complexity_report — per-function length, nesting depth and cyclomatic-style branch count, flagging what is too long or too deep to review confidently • secret_scan — cloud keys, tokens and private-key blocks by line and category; never echoes the matched value • review_checklist — a focused per-language PR checklist • Licence: security_deep_dive (OWASP-oriented pass with the risk class and a fix direction per finding) + get_reviewer_persona (the Senior Reviewer voice as a ready-made system prompt)

USE CASES • Gate an AI coding agent's output before it reaches your main branch • Triage a large PR in seconds to find what deserves human eyes first • Catch a leaked key or a disabled TLS check before it ships

WHO THIS IS FOR Engineering teams using AI coding assistants, solo developers shipping fast, and agent builders who need an automated first-pass reviewer that blocks only on real risk.

WHY MOLTLINE • Every tool tested against known answers before release • One All-Access licence ($19/month, settled in cryptocurrency through NOWPayments) unlocks the premium tools on every Moltline server - send it in the X-Moltline-License header • Support: first response <24h, confirmed defects fixed <30 days


Independently audited — MCPize Verified A (94/100) on this endpoint (https://mcp.moltlinestudio.com/codereview) · Live check result