zephex
CLIGet StartedPricingMCP ToolsCommunityGuidesDocs
←BackSign in
CLIGet StartedPricingMCP ToolsCommunityGuidesDocs
Get started freeSign in
DocsAPIToolsEditorsChangelogHelp

GET STARTED

WelcomeQuickstartSetup videoMCP Q&A (learn)BlogWhat is MCP?Who is Zephex for?Plans & PricingZ-GASAB benchmarkBenchmark chart (live)Changelog

INSTALLATION

Web Terminal tools (plain English)Terminal tools (complete)Connect MCPVS Code Marketplace extensionCLI (no AI agent)CLI init (first run)CLI account & logoutNPX (Recommended)Test Pulse (check test)Test Pulse commandsProject MemorySupply Pulse (supply)Supply Pulse commandsTerminal CLI referenceSlash commands (37 palette)Web Terminal (dashboard)Command CompassCLI commandsCLI in DockerCLI: All editors (one command)CLI: Crush, Hermes, ChatGPT, KiloOAuth & HTTP setupInstall overviewHTTP APISetup WalkthroughHTTP vs stdio

API & KEYS

API Key ManagementKey Naming & FormatAuthenticationKey Dashboard

CONFIGURATION

Universal RequirementsSupported EditorsHow It WorksArchitectureCLAUDE.md TemplateAGENTS.md Template

EDITORS28 guides

Supported EditorsVS CodeVS Code extension (Marketplace)Claude CodeCursorWindsurfJetBrains

PLATFORM

macOSWindowsLinux

TOOLS10 tools

Capabilities OverviewTools OverviewTool FilteringTool Workflowsget_project_contextread_codefind_codecheck_packageexplain_architectureZephex_dev_infocheck_testaudit_headerskeep_thinkingproject_memory

GUIDES

Best PracticesToken EfficiencyUse CasesZephex vs Local MCPZephex vs Context7Zephex vs GitHub MCPZephex vs SmitheryMCP EcosystemMarkdown Access

SUPPORT

Help CenterMCP troubleshootingTeam rolloutFAQConnection IssuesRate LimitsDowntime & ErrorsBillingTier GuidePro & Max guideUsage LimitsUsage Analytics

LEGAL

System StatusTerms (summary)Privacy (summary)Data UseSecurityAuthenticationSecurityData HandlingPrivacy PolicyTerms of Service

Quick Links

API Reference

Complete API documentation

Troubleshooting

Common issues and solutions

Community

Join our Discord community

Plugins

Editor and CLI integrations

Pricing

Free, Pro, and Max plans

Enter
Zephex_devzephex-devzephexzephexhello@zephex.dev
© 2026 Zephex. All systems operational.

Tools

Capabilities

The complete reference of every tool Zephex provides. Each tool is available immediately when your API key is active.

One Endpoint, 10 Tools
Zephex Proxysingle endpointPROJECTget_project_contextcheck_testCODEread_codefind_codeexplain_architecturePACKAGEScheck_packageproject_memoryEXTERNALaudit_headersZephex_dev_infoREASONINGkeep_thinking

The proxy routes each call by tool name to the right backend. Add or remove tools without changing your editor config.

OVERVIEW

Zephex provides 10 tools through a single MCP endpoint. Your AI agent discovers them via tools/listand calls them as needed. No configuration per tool — they're all available on every plan.

CategoryToolsWhat they do
Projectget_project_context, check_testUnderstand projects and plan changes
Coderead_code, find_code, explain_architectureRead, search, and map codebases
Packagescheck_package, project_memoryVerify safety and plan upgrades
Externalaudit_headers, Zephex_dev_infoAudit URLs and access expert knowledge
Reasoningkeep_thinkingStructured debugging and planning
ALL TOOLS

Project Context

get_project_context

Understands your entire project in one call — framework, package manager, scripts, dependencies, entry points, env vars, and monorepo structure.

When to use:

  • First time touching a codebase
  • Before suggesting commands (npm vs bun vs pnpm)
  • Before recommending libraries (check what's already installed)
  • Onboarding to a new repo

Code Reading

read_code

AST-based surgical code extraction. Three modes: symbol (extract by name), file (batch read), outline (structural TOC). Returns signatures and bodies at a fraction of full-file tokens.

When to use:

  • Reading a specific function or class
  • Getting an overview of a large file
  • Batch-reading multiple files within token budget
  • Understanding a symbol's implementation

Code Search

find_code

BM25-ranked code search with AST-aware context. Supports literal, regex, and boolean queries. Returns ranked results with full function/class bodies.

When to use:

  • Finding where a symbol is defined or used
  • Searching for patterns across the codebase
  • Dead code analysis
  • Security auditing (finding hardcoded secrets)

Architecture Analysis

explain_architecture

End-to-end architecture analysis with Mermaid diagrams. Detects entry points, auth flow, data flow, service boundaries, external services, and anti-patterns.

When to use:

  • Understanding how a system is wired
  • Tracing request flows
  • Identifying architectural problems
  • Generating documentation diagrams

Task Scoping

check_test

Give it a plain-English task, get back the minimal file set needed. Returns focus files with roles, reusable utilities, callers at risk, and a suggested approach.

When to use:

  • Before starting any coding task
  • Estimating blast radius of a change
  • Finding which files to read for a bug fix
  • Planning a refactor

Package Safety

check_package

Live registry lookup for any package: version status, supply-chain attack signals, typosquat detection, maintainer changes, weekly downloads, and license.

When to use:

  • Before installing any new dependency
  • Verifying a package name isn't a typosquat
  • Checking if a package is deprecated or abandoned
  • Security review of dependencies

Project Memory

project_memory

Remember and recall project decisions, gotchas, and conventions across sessions via local SQLite — so you never rediscover the same patterns.

When to use:

  • Persisting agent context across sessions
  • Recalling previously discovered project conventions
  • Avoiding repeated investigation of the same patterns
  • Sharing project knowledge between team members

Security Auditing

audit_headers

Wire-level audit of deployed URLs: security headers, SSL/TLS certificates, redirect chains, and cookie flags. Returns an A+ to F grade with fix snippets for your stack.

When to use:

  • Post-deploy security checks
  • SSL certificate expiry monitoring
  • CSP and CORS configuration validation
  • Compliance audits (SOC2, PCI, HIPAA)

Expert Knowledge

Zephex_dev_info

Expert developer knowledge base covering database schemas, security patterns, frontend/backend best practices, authentication flows, and deployment guides.

When to use:

  • Looking up Stripe schema patterns
  • Understanding Supabase RLS policies
  • Getting CSP configuration guidance
  • Learning JWT hardening techniques

Structured Reasoning

keep_thinking

Persistent reasoning sessions that track hypotheses, observations, and conclusions. Prevents circular debugging, detects endless loops, and anchors to the original goal.

When to use:

  • Debugging hard-to-reproduce bugs
  • Planning high-risk changes (auth, billing, schema)
  • Multi-system problems spanning services
  • Trade-off decisions between approaches
COMBINING TOOLS

How tools work together

The real power is in combinations. A typical productive session:

  1. check_test — identify which files matter for your task
  2. read_code — read the specific symbols you need to understand
  3. find_code — search for related patterns or usages
  4. keep_thinking — reason through the approach if it's complex
  5. check_package — verify any new dependency before adding it

See Tool Workflows for detailed examples of multi-tool sessions.