Scripthold

Scripthold — Secure MCP Server for Local Workspaces

Code from the web. Work locally. Recover safely.

Scripthold is a Model Context Protocol (MCP) server that gives web, desktop, and CLI agents controlled access to explicitly authorized local workspaces. It safely handles legacy text encodings, deterministic file changes, source navigation, backups, and optional durable local execution over stdio or authenticated Streamable HTTP.

AI clients see Настройки — not ???? or Íàñòðîéêè.

Start here: Quick start · Find a tool · All documentation · Security policy.

What Scripthold provides

Scripthold 4.x provides 36 tools and 3 guided prompts; see GitHub Releases for the latest published version. Clients upgrading from 3.x should follow the 4.0 migration guide for the unified confirmation tool. In 4.0.1, the source_symbols outline response defaults to a compact tree; pass presentation: "detailed" to retain the earlier files and symbols fields.

• 168 registered text encodings with content-based detection, UTF-32 LE/BE support, and conservative ambiguity handling.
• 100 active Source Intelligence providers for bounded declaration navigation, structural search, selected project relations, and verified context assembly.
• Secure workspace boundaries with canonical-root containment, Windows reparse/junction handling, deterministic traversal, and missing-path validation.
• Approval-bound mutations using preview/apply capabilities, exact fingerprints, conflict checks, staged writes, and truthful partial-state reporting.
• Optional persistent backups with history, compare, audit, restore, explicit garbage collection, pinning, and exact-ID deletion.
• Durable asynchronous tasks with idempotency, persistent state, bounded logs, recovery, locks, and cancellation.
• Reliable long-running read-only calls through optional deferred-operation storage and bounded retained-result retrieval.
• Authenticated Streamable HTTP with loopback defaults, bearer authentication, Host/Origin validation, resource limits, and explicit TLS/proxy requirements for non-loopback exposure.

Scripthold originated from the original mcp-file-tools project, created by Dimitar Grigorov, and retains its GPL-3.0 lineage. See Project Direction.

Quick start

Choose a published binary or build from source, then configure your MCP client to grant access only to the directories you intend to share. HTTP and container deployments have additional setup below.

Use a published release

Download the asset for your platform from the latest release and verify it against checksums.txt. Releases also attach a signed Sigstore build-provenance bundle for the normal GoReleaser assets.

Build from source

git clone https://github.com/zoster81/scripthold.git

cd scripthold go test ./... go build -o scripthold ./cmd/scripthold

The module path is github.com/zoster81/scripthold.

Local stdio client

Pass each authorized directory as a startup argument:

{

"mcpServers": { "scripthold": { "type": "stdio", "command": "C:\\Tools\\scripthold_windows_amd64.exe", "args": ["D:\\Projects"] } } }

Startup roots are authoritative. A roots-capable stdio client may supply dynamic roots only when the process starts without directory arguments.

Verified self-update

Supported standalone installations can opt in to self-update. Stop other Scripthold processes using the same binary before running scripthold update --adopt. Adoption prepares local state without downloading or replacing the executable.

After adoption, run scripthold update to verify and dispatch an available release. Use scripthold update status to confirm completion; only run scripthold update recover when status offers recovery. These actions have different safety and verification requirements.

See the Verified Self-update Guide for supported installations, exact commands, failure handling, and recovery rules.

Streamable HTTP

HTTP requires exactly one bearer-token source. A minimal loopback PowerShell launch is:

$tokenPath = Join-Path $env:TEMP "scripthold.token"

$bytes = New-Object byte[] 32 $rng = [System.Security.Cryptography.RandomNumberGenerator]::Create() try { $rng.GetBytes($bytes) } finally { $rng.Dispose() }

$env:MCP_HTTP_TOKEN_FILE = $tokenPath $env:MCP_HTTP_ADDR = "127.0.0.1:8765" .\scripthold_windows_amd64.exe --transport=streamable-http D:\Projects

The MCP endpoint is http://127.0.0.1:8765/mcp; /healthz and /readyz expose liveness/readiness. Send the token as Authorization: Bearer on every MCP request.

Do not expose HTTP beyond loopback without reading HTTP Security. Non-loopback use requires explicit opt-in and TLS or a trusted proxy boundary.

OpenAI Secure MCP Tunnel

Sanitized PowerShell examples are available under examples/:

• start-local-stdio.ps1
• start-local-http.ps1
• start-openai-tunnel-stdio-plus-local-http.ps1
• start-openai-tunnel-http-plus-local-stdio.ps1

Copy an example outside the checkout before replacing placeholders. Never commit Runtime API keys, Tunnel IDs, bearer tokens, or private state paths.

Container

docker build --build-arg VERSION=dev -t scripthold:dev .

docker run --rm -i \ --read-only \ --cap-drop=ALL \ --security-opt=no-new-privileges \ --tmpfs /tmp:rw,noexec,nosuid,size=64m \ --mount type=bind,source=/absolute/project,target=/data \ scripthold:dev --transport=stdio /data

The image runs as unprivileged UID/GID 10001.

Tool groups

TOOLS.md is the detailed public reference for schemas, parameters, examples, limits, and error behavior. The table below describes the 4.x catalog (36 tools). Clients upgrading from 3.3.6 must migrate the eight former apply commands to confirm_verified_operation.

Source Intelligence

Source Intelligence helps agents navigate source code without running it or requiring an external compiler, parser, or language server. Use source_symbols to map project files with digest presentation: "compact", browse a tree-structured file outline by default (or request presentation: "detailed" for the previous outline JSON structure), and inspect a selected symbol; use source_query for structural search, supported project relationships, dependency graphs, and verified context. Use grep_text_files when looking for arbitrary text or code behavior that is not represented by structural symbols.

Unsupported or ambiguous relationships are rejected rather than guessed. Markdown files have their own document tools. Check the generated Language Capabilities matrix for available source providers and analysis coverage.

Markdown documents

Markdown files use dedicated tools rather than Source Intelligence providers. Read a document with markdown_read, or inspect links and relationships across files with the read-only markdown_workspace tool.

To change a document, prepare a preview with markdown_edit (or markdown_create for a new file), review the proposed content, and pass its previewId to confirm_verified_operation. Unsafe, conflicting, or stale changes are rejected instead of being silently rewritten.

For examples and the exact operation forms, see the Markdown tool reference. For Markdown semantics, supported encodings, and the Marksplice integration boundary, see Markdown Intelligence.

Safety model

• File access is limited to explicitly authorized roots after canonical path validation.
• Symlink, junction, reparse-point, alias, and missing-path escapes fail closed.
• Encoding detection uses bytes and decoded-content evidence, never filenames or extensions.
• Mutations st