Any ScriptSAFE
An MCP server that exposes arbitrary CLI tools and shell scripts as MCP Tools
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://mseep.ai/app/izumin5210-any-script-mcp)
An MCP server that exposes arbitrary CLI tools and shell scripts as MCP Tools
日本語版
Overview
An MCP server that publishes commands defined in YAML files as MCP Tools. By defining tools in a configuration file, you can execute arbitrary shell scripts from MCP clients.
Installation
npx
Claude Code:
$ claude mcp add any-script \ -s user \ -- npx any-script-mcp
json:
{
"mcpServers": {
"any-script": {
"command": "npx",
"args": ["any-script-mcp"]
}
}
}Configuration
Create a configuration file at $XDG_CONFIG_HOME/any-script-mcp/config.yaml (typically ~/.config/any-script-mcp/config.yaml).
You can also specify custom configuration file paths using the ANY_SCRIPT_MCP_CONFIG environment variable:
# Single configuration file $ ANY_SCRIPT_MCP_CONFIG=/path/to/custom/config.yaml npx any-script-mcp # Multiple configuration files (Unix/macOS - separated by colon) $ ANY_SCRIPT_MCP_CONFIG=/path/to/custom.yaml:$XDG_CONFIG_HOME/any-script-mcp/config.yaml npx any-script-mcp # Multiple configuration files (Windows - separated by semicolon) $ ANY_SCRIPT_MCP_CONFIG=C:\path\to\custom.yaml;%APPDATA%\any-script-mcp\config.yaml npx any-script-mcp
When multiple configuration files are specified:
- All tools from all files are merged into a single collection
- If the same tool name appears in multiple files, the first occurrence takes precedence
- At least one valid configuration file must be successfully loaded
- This is useful for separating common tools from project-specific or personal customizations
Testing Your Configuration
You can test your configuration using the MCP Inspector:
$ npx @modelcontextprotocol/inspector npx any-script-mcp
This will open a web inter
c40770e3dff7OBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add any-script-mcp -- npx -y [email protected]
{
"mcpServers": {
"any-script-mcp": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (28)
28 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
all_types | read | Test all input types |
boolean_test | read | Test boolean inputs |
compatibility_test | read | Test backward compatibility |
complex_json_test | read | Test INPUTS_JSON with complex data |
custom_shell_test | read | Test custom shell command |
default_shell_test | read | Test default shell from config |
default_timeout_test | read | Test default timeout |
echo_test | read | Test echo |
error_test | read | Test execution errors |
existing_tool | read | Existing tool |
hyphen_test | read | Test hyphenated names |
json_test | read | Test INPUTS_JSON environment variable |
json_type_test | read | Test INPUTS_JSON with types |
multi_test | read | Test multiple inputs |
multiline_test | read | Test multiline scripts |
no_inputs | read | Tool without inputs |
node_json_test | read | Test INPUTS_JSON with Node.js |
node_test | read | Test Node.js execution |
optional_test | read | Test optional parameters |
placeholder_test | read | Test {0} placeholder replacement |
python_json_test | read | Test INPUTS_JSON with Python |
required_test | read | Test required parameters |
sh_test | read | Test sh shell |
timeout_success_test | read | Test successful execution with timeout |
timeout_test | read | Test custom timeout |
tool_one | read | First tool |
tool_two | read | Second tool |
type_test | read | Test parameter types |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (1)
@modelcontextprotocol/sdk, execa, xdg-basedir, yaml, zod, vitest, zod-to-json-schema
Gates applied: no_behavioural_pass.
c40770e3dff7full audit observations/trust-audit/mcp-server/izumin5210__any-script.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | c40770e3dff7 | SAFE | B | 89 | first audit |
Questions
What is the Any Script MCP server?
An MCP server that exposes arbitrary CLI tools and shell scripts as MCP Tools
What tools does Any Script expose?
28 in total: 28 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Any Script safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Any Script need?
No credential environment variables were found in its source, so it appears to need none.
How does Any Script run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as any-script-mcp at 0.3.0.
How current is this page?
The grade is for one exact copy of the source (c40770e3dff7), read on 2026-10-08. The repository is watched and re-audited when it changes.