About mcp-portal
mcp-portal is a stdio MCP server that delegates bounded file reading and reference-based code generation to the Cursor CLI. It is intended for agents and verifies quoted evidence from selected files before returning it.
Key features
- 01Read selected files with cited quotes
- 02Generate code from a reference and spec
- 03Write generated code to a target file
- 04Apply model-selection policy
- 05Restrict file access and verify citations
Use cases
- 01Answer questions using a bounded set of files
- 02Generate boilerplate matching a reference file
- 03Delegate code writing through Cursor CLI
- 04Verify file-based answers with source quotes
MCP support
It exposes `bulk_read`, `code_write`, and `status` tools to MCP hosts. The server delegates work to the locally configured Cursor CLI and verifies citations for bulk reads.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
uvx mcp-portalMCP_PORTAL_HOMEOptional cache and evidence directory (default ~/.cache/mcp-portal)MCP_PORTAL_CLIOptional path to cursor-agent or agent executableLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
3Saved tool list · updated 3w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| root | string | No | |
| model | string | No | |
| paths | array | Yes | |
| question | string | Yes |
Details
From the project READMEmcp-portal
mcp-portal is a stdio Model Context Protocol server that lets a frontier agent (Claude Code, Codex, Cursor, or any MCP host) delegate two jobs to the Cursor CLI on its own quota: bounded bulk_read (read explicitly selected files, answer with verified quotes) and code_write (generate boilerplate from a reference file + spec; the server writes the target file). Python stdlib only—no Node runtime and no MCP SDK dependency.
On 2026-09-08, composer-2.5-fast generated roughly 5× faster than a frontier model on the same brief (line-rate measurement). Cursor quota is separate from the host model's.
Quick start
uvx mcp-portal
Also available as pipx install mcp-portal / pip install mcp-portal, and listed in the
MCP Registry as io.github.apollion69/mcp-portal.
To run the development head instead of the release:
uvx --from git+https://github.com/apollion69/mcp-portal mcp-portal
Requirements: Python 3.10+, the Cursor CLI (cursor-agent) installed and logged in.
Doctor (CLI inventory, no model call):
mcp-portal-doctor
Configure per host
Claude Code
claude mcp add --scope user mcp-portal -- uvx mcp-portal
Codex (~/.codex/config.toml)
[mcp_servers.mcp-portal]
command = "uvx"
args = ["mcp-portal"]
tool_timeout_sec = 150
Cursor (~/.cursor/mcp.json)
{
"mcpServers": {
"mcp-portal": {
"command": "uvx",
"args": ["mcp-portal"]
}
}
}
VS Code (.vscode/mcp.json, servers key)
{
"servers": {
"mcp-portal": {
"type": "stdio",
"command": "uvx",
"args": ["mcp-portal"]
}
}
}
Generic mcpServers JSON
{
"mcpServers": {
"mcp-portal": {
"command": "uvx",
"args": ["mcp-portal"]
}
}
}
Environment (optional):
| Variable | Purpose |
|---|---|
MCP_PORTAL_HOME | Cache, receipts, evidence (default ~/.cache/mcp-portal) |
MCP_PORTAL_CLI | Path to cursor-agent / agent |
Tools
bulk_read
| Argument | Required | Description |
|---|---|---|
paths | yes | 1–16 file paths (relative to root or absolute) |
question | yes | Question answered only from those files |
root | no | Common root; default = longest common parent of paths |
model | no | Override model; policy applies when omitted |
Returns status, run_id, answer.findings[] (file, start, end, quote, fact), gaps[], metrics, model_decision.
code_write
| Argument | Required | Description |
|---|---|---|
spec | yes | What to generate |
reference_path | yes | Style/context reference file |
target_path | no | If set, server writes this path |
model | no | Override model |
Returns generated code, optional bytes_written, run_id, metrics.
status
No arguments. Returns CLI path, auth hint, default model, policy summary, cache location, receipt counters.
Model policy
Shipped in model-policy.json (package data). Defaults:
- Prefer Cursor-native models (
composer-2.5, thencursor-grok-*) - Strip
-fastsuffixes (never auto-select fast variants) - Other vendors only when explicitly requested and listed by
cursor-agent --list-models
Override by editing model-policy.json in the installed package or setting policy fields via a custom file at MCP_PORTAL_HOME (future) — today, replace the package file or patch preferred in your fork. Each tool result includes model_decision.reason (default_preferred, fast_suffix_stripped, cursor_native_explicit, explicit_other_vendor, requested_unavailable_fallback).
How it works
- Authorize — Server reads only listed paths; blocks credential-like paths and secret patterns.
- Manifest — Request JSON includes per-file SHA-256 hashes.
- Isolate — Cursor CLI runs with fresh
CURSOR_CONFIG_DIR, deny-all permissions,--mode ask, sandbox enabled. - Verify — Every
quoteinbulk_readanswers must appear verbatim in the cited line range; bad citations are dropped or fail closed. - Evidence — Per-run directory under
MCP_PORTAL_HOME/runs/<run_id>/with manifest (hashes, metrics; not full source). - Budgets — 16 files, 128 KiB combined input, 90s timeout, bounded stdio frames.
Windows
On Windows, the delegate uses a local Cursor CLI run when either:
MCP_PORTAL_CLIpoints at an executable (including test stubs), orcursor-agent/agentis found onPATHand is a real file.
Otherwise it falls back to the wsl.exe bridge into Ubuntu/WSL (python3 -m mcp_portal.delegate --worker). Force either mode with MCP_PORTAL_BACKEND=local or MCP_PORTAL_BACKEND=wsl.
- MCP config can use native
uvx mcp-portalwhen the CLI is on PATH, orwsl.exe+uvx mcp-portalwhen it is not - Helpers in
clients/windows/(delegate.ps1,parse_read.ps1) MCP_PORTAL_WORKERoverrides the default WSL worker commandMCP_PORTAL_WSL_CDsets the WSL working directory (default~)
Optional Claude Code routing hook
Install read gate + skill (generic, transactional):
python3 -m mcp_portal.install_router prepare --client claude --python python3 \
--state-root ~/.cache/mcp-portal/router-tx --shell bash --command-shell bash
# then apply with the printed transaction id
See docs/skills/cursor-bulk-reader/SKILL.md for agent-facing guidance. The router blocks or warns on large full-file reads (>350 lines or >128 KiB) and points agents at bulk_read.
Repo-level MCP registration helper:
python3 -m mcp_portal.install plan
python3 -m mcp_portal.install apply --target claude-mcp
Security
See SECURITY.md. Summary: you choose which files leave the machine; the CLI runs read-only with tools denied; quotes are verified server-side. Not a substitute for secret hygiene.
Related projects
Several Node-based bridges expose Cursor via MCP (different tradeoffs: SDK/Node stack, varying isolation and verification):
- lipey1/cursor-agent-mcp
- andreilungeanu/cursor-delegate-mcp
- sailay1996/cursor-agent-mcp
- ai-nuke/cursor-agent-mcp
- JaimeJunr/cursor-mcp-bridge
mcp-portal focuses on stdlib Python, hash-pinned manifests, quote verification, server-side writes for code_write, model policy, and WSL-first Windows support.
License
MIT — see LICENSE.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.