Fastest: npx -y zephex setup --continue (browser OAuth, writes stdio). Global ~/.continue/config.yaml or project .continue/mcpServers/zephex.json. MCP only works in Continue agent mode.
Official Continue MCP documentation: Continue MCP docs
Without MCP, your agent guesses project layout and misses supply-chain risk. Zephex connects one hosted endpoint so every session gets the same ten tools — no per-machine npm installs, no version drift across the team.
HTTP and stdio setups both need a key from your Zephex dashboard. OAuth-only flows (ChatGPT, Claude.ai web) sign you in in the browser instead — skip this section for those.
Matches the published CLI (mcp-proxy/src/commands/setup.ts). One command signs you in, writes the correct transport, and verifies 10 tools.
Account teardown: logout vs disconnect · Connect MCP walkthrough
Search docs for “where is my MCP file” — the answer is always: run list first, then open the path it prints.
Documented paths: ~/.continue/config.yaml or <project>/.continue/mcpServers/zephex.json
Continue wizard appends stdio YAML/JSON.
Runs the same code as mcp-proxy/src/commands/setup.ts — OAuth in browser, creates a CLI key, writes your editor config, verifies tools.
npx -y zephex setup --continueSkip browser OAuth — paste a key from zephex.dev/dashboard/keys. Must start with mcp_prod_, mcp_dev_, or mcp_sk_.
npx -y zephex setup --continue --api-key mcp_prod_your-key-hereIf you edit by hand after setup: Transport: stdio (see ~/.continue/config.yaml or <project>/.continue/mcpServers/zephex.json).
{ "mcpServers": { "zephex": { "command": "npx", "args": ["-y", "zephex"], "env": { "ZEPHEX_API_KEY": "mcp_sk_your_key_here" } } }}After CLI install, fully restart the app if tools do not appear. Manual JSON/TOML blocks below are equivalent — use them when CLI commands are unavailable.
Full comparison: HTTP vs stdio · npx zephex reference
Setup writes stdio (npx -y zephex) so get_project_context and read_code can read your workspace. Manual streamable-http YAML is optional only.
Replace mcp_sk_your_key_here with your key from Dashboard → API Keys. Copy the full key once at creation — paste into Authorization: Bearer … for HTTP configs, or into ZEPHEX_API_KEY for stdio/npx configs.
Appended under mcpServers: in ~/.continue/config.yaml — command npx, args, env.
mcpServers: - name: zephex command: npx args: - "-y" - "zephex" env: ZEPHEX_API_KEY: "mcp_sk_your_key_here"JSON drop at workspace root — matches setup.ts writeContinueConfig project branch.
{ "mcpServers": { "zephex": { "command": "npx", "args": ["-y", "zephex"], "env": { "ZEPHEX_API_KEY": "mcp_sk_your_key_here" } } }}Not written by setup — use only if you cannot run npx stdio and will pass github: URLs.
name: Zephex MCPversion: 0.0.1schema: v1mcpServers: - name: zephex type: streamable-http url: https://zephex.dev/mcp headers: Authorization: "Bearer mcp_sk_your_key_here"Tip
MCP tools are not available in non-agent Continue modes — switch to agent before testing.
Note
Remove old streamable-http or command/args blocks before switching transports — duplicate zephex servers confuse Continue.
Run these in a terminal when the editor UI is unclear — catches stale npm, wrong transport, and project shadows.
npx -y zephex@latest listnpx -y zephex@latest doctornpx -y zephex@latest repair# Fully quit the editor (Cmd+Q / Alt+F4), reopen, start a new agent sessionAfter saving your config, confirm Zephex is connected before you rely on it in real work.
Questions people ask when Continue does not show Zephex tools — indexed for docs search.
Continue Zephex not in agent mode
Switch Continue to agent mode — MCP servers do not load in other modes.
Continue YAML vs JSON for Zephex
Setup uses stdio YAML globally or stdio JSON under .continue/mcpServers/. Manual HTTP is optional only.
You do not call tools yourself — ask your agent in plain language. Try these once Zephex is connected:
“Continue agent mode only: check_package on langchain before pip install in this notebook repo.”
MCP tools are available in agent mode per Continue docs.
“get_project_context on the workspace with .continue/mcpServers/zephex-mcp.yaml.”
YAML block server attaches to the same repo root.
“check_test: add eval harness for our RAG pipeline — which modules to touch?”
ML-ish repos benefit from explicit scope before edits.
“find_code vector store client initialization.”
Search when notebooks hide structure.
“read_code the embedding batch function — outline if the file is huge.”
AST read for Python ML code.
“Zephex_dev_info: pgvector vs pinecone tradeoffs for our scale.”
Expert comparison without leaving Continue.
Switch Continue to agent mode — MCP is not available in every UI mode. YAML servers under .continue/mcpServers/ apply to this workspace.
These situations usually mean the setup cannot work until you fix the underlying issue:
Server not listed
Global: ~/.continue/config.yaml with `- name: zephex`. Project: .continue/mcpServers/zephex.json. Run list.
0 tools
Agent mode + stdio command/args/env. Run repair if doctor flags unpinned zephex.
Auth failed (manual HTTP)
Authorization: "Bearer YOUR_KEY" only if you chose streamable-http.
Wrong mode
Switch Continue to agent mode.
npx not found in Continue
npx -y zephex@latest repair + system Node on PATH; quit VS Code fully.
Continue agent mode + stdio setup (YAML or project JSON) — ten tools:
get_project_context
Reads your project structure, dependencies, scripts, env vars, and framework markers in one call. Replaces manually opening package.json, tsconfig, and multiple config files at the start of every session.
read_code
AST-based code extraction: pass a symbol name and get the implementation without reading entire files. Supports symbol lookup, batched file reads, and structural outlines for large files.
find_code
Ranked search across the repo for definitions, usages, and patterns. Faster than blind grep when the agent does not know where a symbol lives.
check_package
Live registry lookup for npm, PyPI, Cargo, and Go modules. Surfaces typosquat risk, maintainer changes, and suspicious version jumps before you run install.
explain_architecture
Generates Mermaid diagrams for auth flows, service boundaries, and module dependencies so the agent reasons about structure instead of guessing.
check_test
Turns a task description into the smallest file set to read or edit, with risk ratings and caller impact notes.
audit_headers
Grades a deployed URL for CSP, HSTS, TLS, cookies, and redirects. Returns fix snippets for common hosts (Vercel, Cloudflare, Nginx).
keep_thinking
Structured multi-step debugging: tracks hypotheses and conclusions so long investigations do not loop.
Zephex_dev_info
Expert patterns for authentication, databases, frontend frameworks, deployment, and mobile stacks when the agent needs vetted guidance.
project_memory
Persists decisions, gotchas, and conventions per project in ~/.zephex/memory (SQLite FTS5). recall before unfamiliar areas; remember after discoveries. Local stdio only on npx zephex.