fauxnix
fauxnix
Run Linux-style commands on Windows — natively, deterministically, with no VM and no WSL.
fauxnix is a bash→PowerShell translation layer built for AI agents. Your agent keeps writing the
bash it already knows (ls -la | grep foo, find . -name '*.ts' | wc -l, kill -9 1234), and
fauxnix deterministically translates each command into PowerShell, executes it natively, and hands
back output that looks like GNU/Linux: ls -l columns, bash-style error messages, coreutils exit
codes, UTF-8/GBK handled automatically.
npm install -g fauxnix-cli # then point any MCP harness at `fauxnix mcp`$ fauxnix "ls -la src | head -2"
-rw-r--r-- 1 me me 1204 Aug 16 09:12 ast.ts
-rw-r--r-- 1 me me 8192 Aug 16 09:12 cli.ts
$ fauxnix "cat nope.txt"
cat: nope.txt: No such file or directory # not a PowerShell stack traceMeasured: your model is probably worse at PowerShell than you think
Same model (DeepSeek-V4-Pro), same 5 tasks, three execution modes on one Windows machine —
full data in docs/benchmark-deepseek-v4-pro.md and
[`docs/benchmark-ark-models.md](docs/benchmark-ark-models.md):
PowerShell | fauxnix | Git Bash | |
tool calls / unexpected errors | 14 / 9 | 7 / 0 | 4 / 0 |
time (T1–T4) | 163s | 66s | 57s |
Across 7 models on the Volcano Ark Coding Plan, the PowerShell-vs-fauxnix gap held for every model tested — worst case (kimi-k2-thinking): 3.1× slower with 24 error events writing PowerShell vs zero errors through fauxnix. fauxnix lands within ~15% of the real-bash ceiling with no bash toolchain installed.
Related MCP server: wmux
Why
LLM agents are dramatically better at bash than at PowerShell — bash dominates training data, so
models on Windows often produce "looks right, doesn't run" commands (wrong quoting, curl that
isn't curl, mojibake from codepage mismatches, inscrutable CategoryInfo error dumps). Existing
solutions are either a full VM (WSL — heavy, wrong filesystem, separate environment) or plain
shell wrappers (still PowerShell underneath).
fauxnix takes the third road: translate, don't emulate. A large, high-value subset of the Linux command line — file ops, text processing, process management, archives, networking basics — maps cleanly onto PowerShell + .NET. fauxnix implements that subset faithfully and fails loudly and helpfully on what it can't translate, so the agent never gets silently-wrong results.
Install
npm install -g fauxnix-cliOr from source:
git clone https://github.com/20000419/fauxnix && cd fauxnix && npm install -g .npm package name is
fauxnix-cli(thefauxnixname on npm belongs to an unrelated 2015 websocket library); the installed command is stillfauxnix.
Requires: Windows with PowerShell 5.1+ (built-in) and Node.js ≥ 18.
Quick start
# one-off commands
fauxnix "ls -la"
fauxnix "grep -rn TODO src | wc -l"
fauxnix "cat log.txt | grep -i error | sort | uniq -c"
# see what a command becomes (great for debugging / learning PS)
fauxnix translate "find . -name '*.log' -mtime +7 -delete"
# check your environment
fauxnix check
# run the MCP stdio server (what agent harnesses connect to)
fauxnix mcpUnknown commands (git, node, npm, python, cargo, gh, docker, ...) are passed through natively with argv-style quoting — no string re-parsing, no quoting bugs.
Use with your agent harness
fauxnix ships an MCP stdio server exposing a bash tool (plus fauxnix_translate and
fauxnix_session). Point any MCP-capable harness at it:
Claude Code
claude mcp add fauxnix -- fauxnix mcpCodex (~/.codex/config.toml or codex mcp add fauxnix -- fauxnix mcp)
[mcp_servers.fauxnix]
command = "fauxnix"
args = ["mcp"]Note: in non-interactive codex exec mode, MCP tool calls are auto-denied by
the approval layer; pass --dangerously-bypass-approvals-and-sandbox (or run
interactively and approve once).
OpenCode (opencode.json)
{
"mcp": {
"fauxnix": { "type": "local", "command": ["fauxnix", "mcp"] }
}
}Kimi Code — unlike the others, MCP servers live in a JSON file, not the
TOML config: ~/.kimi-code/mcp.json
{
"mcpServers": {
"fauxnix": { "command": "fauxnix", "args": ["mcp"] }
}
}Any MCP client — stdio server: fauxnix mcp. The tool name is bash (override with
FAUXNIX_TOOL_NAME). Tool description already teaches the model the supported subset, so no
system-prompt changes are required.
The MCP session persists cwd, environment variables, export/unset and cd -/OLDPWD across
tool calls — it behaves like a logged-in shell, not a stateless exec.
What's translated
~105 commands, all output-matched against real GNU coreutils on Windows (Git Bash) during development:
files:
ls cp mv rm mkdir rmdir touch mktemp ln readlink realpath basename dirname stat file du df find chmod chown difftext filters:
grep egrep sed awk sort uniq cut tr— sed/awk scripts are parsed at translate time (unsupported constructs throw named errors, never silently misbehave)text I/O:
echo printf cat head tail wc tee nl tac md5sum sha1sum sha256sum base64 seq yes xargsshell/system:
cd pwd export unset env printenv ps kill pkill pgrep sleep which type whoami id groups date uname hostname uptime free nproc clear true false test [ [[ : pushd popd dirs sudo timeout man history less more source . eval exit alias setnetwork:
curl wget ping netstat ss ip ifconfig nslookup dig hostarchives:
tar gzip gunzip zcat zip unzip
Plus shell syntax: pipes, && / || / ;, redirections (> >> 2> 2>&1 < &>, /dev/null),
quoting, $VAR $(...) command substitution, VAR=x cmd prefixes, ~ expansion, and
POSIX-style path normalization (/tmp, /d/foo → D:\foo).
Exit codes follow bash conventions: 0 ok, 1 fail, 2 usage/serious, 127 command not found, 124 timeout.
How it works
bash command ──parser──▶ AST ──translator──▶ PowerShell script ──executor──▶ powershell.exe
│
agent ◀── GNU-style output, bash-style errors ◀── decoder (UTF-8 → GBK fallback) ◀┘Deterministic translation, zero LLM calls at runtime.
Each command maps to a generator that emits a self-contained PowerShell block honoring the "Fauxnix contract": string-per-line stdout,
[Console]::Error.WriteLinefor bash-style stderr,$script:fx_exitfor exit codes,$inputfor stdin.The executor wraps every script with UTF-8 enforcement (
[Console]::OutputEncoding,$OutputEncoding,chcp 65001), decodes output as strict-UTF-8 with a GBK(936) fallback for legacy native tools, strips CLIXML serialization and PowerShell noise from stderr, and rewrites common PowerShell errors (including zh-CN locale messages) into bash phrasing.Scripts run via
-EncodedCommand(UTF-16LE) and transparently fall back to a temp.ps1file when the 32 KB command-line limit would be exceeded.
Known deviations (honest list)
fauxnix optimizes for the commands agents actually run. Documented deviations:
X=1standalone assignments followexportsemantics (one session-wide environment; bash's shell-var vs exported-var distinction does not exist), and a same-segment prefix is visible to$VARinside the command's own words (Z=in [[ $Z == in ]]is true here, false in bash where word expansion precedes the temporary environment).yesis capped at 65,536 lines — PS 5.1 pipelines cannot signal upstream producers to stop, so an unboundedyes | headwould hang.tail -f,source,eval,alias, heredocs, backticks, shell control flow (if/for/while) and background&are rejected with actionable error messages instead of misbehaving.chmodmaps only the read-only bit; exec bits are no-ops on Windows.chownis a silent no-op (as in Git Bash).ps auxcolumns are approximations (no per-process CPU% accounting, USER shows?).gzip -c/pipeline stdin is text-faithful, not byte-faithful; file-modegzip fis byte-exact.A pipeline producing exactly one line, piped into
wc -l, counts that line (bash would count 0 if the producer omitted the trailing newline).printf 'x' | md5sumstays byte-exact.sed/awksupport the common subset; hold-space, labels, arrays, loops throw named "not supported" errors at translate time.curl/wgetrefuse loopback/private/reserved addresses (localhost, 127.x, ::1, 10.x, 172.16–31.x, 192.168.x, 169.254.x) as a safety default for agent-driven HTTP.Native-tool pipelines vs encoding: PS 5.1 has a single console-encoding knob, so piping localized admin tools (ipconfig, tasklist — GBK on zh-CN) and UTF-8-native dev tools (node, curl) cannot both decode cleanly mid-pipeline. Default favors UTF-8 dev tools; set
FAUXNIX_NATIVE_ENCODING=ansiwhen your agents grep Chinese output of native Windows admin tools. File reads are always sniffed per file (UTF-8 strict → GBK fallback), so grep/sed/awk over GBK files works in either mode — unlike Git Bash, which only matches the encoding its locale assumes.
Development
npm install
npm test # unit + real-PowerShell integration suite (Windows only, auto-skipped elsewhere)
npm run build
npx tsx scratch/run.mjs "any bash command" # quick live checkArchitecture map: src/parser.ts (bash subset → AST) · src/translator.ts (AST → PowerShell +
executor wrapper) · src/executor.ts (spawn, redirects, session persistence) ·
src/commands/*.ts (per-command generators) · src/mcp.ts (MCP server) · src/cli.ts.
License
MIT © 20000419
This server cannot be installed
Maintenance
Related MCP Servers
- AlicenseBqualityDmaintenanceHigh-performance MCP server giving AI agents advanced filesystem and automation capabilities on Windows, with 26 tools across file I/O, search, Git, process management, and more.262MIT
- Alicense-qualityAmaintenanceA native Windows terminal multiplexer with MCP bridge for AI agents, enabling browser automation, multi-agent coordination, and terminal control.343MIT
- Alicense-qualityAmaintenanceA Windows-first, dual-mode MCP bridge that delegates bounded work to Hermes Agent and a local Qwen model via an OpenAI-compatible endpoint, providing restricted batch or trusted full execution modes.MIT
Related MCP Connectors
Package intelligence MCP for AI agents — 22 tools, 19 ecosystems, AGPL SDK, free.
Deterministic reasoning stack for AI agents: simulate, decide & compute, plus cross-domain tools.
Deterministic AI agent microtools, no accounts/API keys. fetch_extract: 98% token cut. 38 tools.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/20000419/fauxnix'
If you have feedback or need assistance with the MCP directory API, please join our Discord server