Open-source · local-first · model-agnostic

Silk Code The AI coding platform you own.

One open harness that understands your repository, edits code, runs your tests, and manages git — on your machine, with any model you choose: DeepSeek, Qwen, Kimi, GLM, MiniMax, Cloudflare Workers AI, or weights running on your own GPU. No subscription. No lock-in. No code leaving your disk unless you say so.

The future of development is local.
The environment belongs to the developer. The model is replaceable.

Why Silk Code exists

Your development infrastructure should be yours

AI coding tools today rent you back your own workflow: your code flows through someone else's servers, into someone else's model, under someone else's rules. Silk Code inverts that. The harness is open source, runs where your code lives, and treats every model — cloud or local — as a swappable component.

Your machine

The runtime, the sessions, the checkpoints, and the project memory live on your disk. Pull the plug on the internet and keep coding. Local models work offline, start to finish.

Your models

Any OpenAI-compatible endpoint, native Ollama, vLLM, LM Studio. Switch mid-session. When a better model ships tomorrow, you keep your entire setup and swap one line of config.

Your rules

Permissions, approvals, project conventions in SILKCODE.md, sandboxing — the agent operates under your governance. MIT-licensed, single runtime dependency, auditable end to end.

The shift

Stop renting. Start owning.

The rented way

Closed AI IDEs & subscriptions

  • Your code is uploaded to vendor clouds for indexing and inference
  • One vendor's model, one vendor's price — take it or leave it
  • Features, limits, and privacy terms change without asking you
  • Your workflow dies the day the subscription lapses or the startup pivots

The Silk Code way

Infrastructure you own

  • Code stays on disk; local models mean zero data leaves the machine
  • Any model, any provider — cloud for power, local for privacy, per session
  • Open source, MIT license — fork it, audit it, extend it, it's yours
  • Sessions, memory, and checkpoints are files you control forever

The loom

One harness, weaving models into real work

Silk Code sits between you and any model, giving it the tools, context, and guardrails to do real engineering — not just autocomplete.

Stage 01

Understand

Every session opens with a repo map, your SILKCODE.md rules, project memory, and installed skills. The model knows your codebase before its first tool call.

Stage 02

Act

Search, read, and edit files; run commands; call MCP tools. Writes and risky commands are approval-gated — rm -rf, push, and merge always ask first.

Stage 03

Verify

run_tests auto-detects pytest, npm, cargo, go, and flutter. The agent runs your suite and reads the failures before claiming success.

Stage 04

Sign & ship

Checkpoints snapshot every file before it's touched — one /revert undoes a turn. Commits carry co-author and model provenance. Push when you say so.

Any model, cloud or local

The model is a component. Swap it freely.

One provider layer speaks to every OpenAI-compatible endpoint, with native Ollama support and local-model discovery. Run parallel sessions on different models and compare — the harness doesn't care which brain you plug in.

DeepSeekQwenKimiGLMMiniMax Cloudflare Workers AIOpenRouter OllamavLLMLM Studio any OpenAI-compatible endpoint

Gold, dashed = runs on your own hardware. --model auto picks whatever you have: a running local server first, then any cloud provider with a key. Onboard anything else with silkcode models add — or pull local models Ollama-style with silkcode models pull qwen2.5-coder.

Quick start

Own your setup in three minutes

Requires Python 3.10 or newer (macOS ships 3.9 — grab a current Python from python.org first) and git. Keep Silk Code in its own virtual environment, then add a shell alias so the silkcode command is available from every project without activating the environment first.

01

Install or update Silk Code recommended

# install or update Silk Code
if [ -d "$HOME/SilkCode/.git" ]; then
    git -C "$HOME/SilkCode" pull
else
    git clone https://github.com/RupertCloud/SilkCode.git "$HOME/SilkCode"
fi

cd "$HOME/SilkCode"

# create the virtual environment if needed
[ -d .venv ] || python3 -m venv .venv

# install or update Silk Code
.venv/bin/python -m pip install -e .

# add the global command if it is not already configured
grep -q 'alias silkcode=' ~/.zshrc 2>/dev/null || \
    echo 'alias silkcode="$HOME/SilkCode/.venv/bin/silkcode"' >> ~/.zshrc

source ~/.zshrc

silkcode --help

This command block is safe to rerun: it updates an existing checkout, reuses its virtual environment, and avoids adding duplicate aliases. The alias invokes Silk Code directly inside its environment, so you do not need to activate .venv to use it.

02

Connect a model — either side works

# cloud: DeepSeek (or qwen / kimi / glm / minimax / openrouter)
export DEEPSEEK_API_KEY=[redacted]

# or local & private: install Ollama from ollama.com, then
silkcode models pull qwen2.5-coder

Pick a model with tool-calling support — the agent works by calling tools. qwen2.5-coder is the reliable local choice.

03

Point it at a project

cd ~/Projects/my-project
silkcode gui .                        # browser GUI at 127.0.0.1:8377

silkcode gui ~/Projects/my-project    # or pass the project path directly
silkcode ~/Projects/my-project        # terminal REPL

silkcode gui ~/SilkCode opens the Silk Code source repository itself. To work on another codebase, pass that project's directory instead.

Then just ask: "Build a small Flask API with tests, and make sure the tests pass." Approve its actions as prompts appear; switch to --mode agent once you trust it.

The harness

Everything between you and the model

Multi-session GUI

Project explorer, streaming chat, agent activity timeline, diff viewer. Open parallel sessions with different models; a busy session keeps working while you use another.

Repository intelligence

Every session starts with a repo map, your SILKCODE.md project rules, accumulated project memory, and installed skills — the model knows the codebase before its first tool call.

Permissions with teeth

Commands are risk-classified. Reads run free, writes and installs ask, rm -rf / push / merge always require approval — unless you explicitly grant them for the session.

Checkpoints & revert

Files are snapshotted before the agent touches them. One click (or /revert) undoes a whole turn.

Verifies its own work

run_tests auto-detects pytest, npm, cargo, go, and flutter — the agent runs your suite and reads the failures before claiming success.

Long sessions that survive

Context compaction trims old tool output and turns as you approach the model's window — sessions roll on instead of dying, and are resumable from GUI or CLI.

MCP client

Connect any Model Context Protocol server — its tools become agent tools, approval-gated like everything else.

Remote sandboxes

Run the agent's commands in a disposable container instead of your machine: self-hosted with one command, or on Cloudflare Sandboxes via the bundled Worker.

Built-in evals

silkcode benchmark runs real coding tasks end-to-end per model — with an A/B mode that isolates what the harness itself contributes.

GitHub

Authorize like an app, ship like a team

Sign in with GitHub — install the Silk Code app, approve a code in your browser, done. No tokens to create or paste; short-lived credentials, scoped to the repos you chose, refreshed automatically.

silkcode connect githubsign in (or verify a $GITHUB_TOKEN)
/push · ⇧ Pushpush when you say so
--auto-pushpush automatically after each turn
github_create_pragent opens draft PRs, reads issues, merges on approval

And the harness signs its work. Agent-made commits register Silk Code as co-author — you stay the author, the record stays honest:

Fix login redirect after session refresh

Co-Authored-By: Silk Code <agent@silkcode.dev>
X-Silk-Model: deepseek/deepseek-chat
X-Silk-Session: 42

git log --grep=X-Silk-Model forever answers "which commits did the agent write, with which model?"

Command line

The whole harness from the terminal

silkcode [path]interactive REPL — /model, /mode, /diff, /revert, /push, /usage
silkcode -p "…"one-shot, scriptable: run a single request and exit
silkcode reviewAI review of your uncommitted changes
silkcode testrun the project's tests, framework auto-detected
silkcode modelslist, add, pull, or set default models
silkcode sessions · resumecontinue any session — including ones started in the GUI
silkcode mcp · sandbox · benchmarktool servers, remote execution, model evals

The contract

What Silk Code guarantees you

The developer's oath

  • MIT licensed — yours to fork, audit, and extend, forever
  • One dependency — httpx; the whole harness is readable in a weekend
  • Zero telemetry — nothing phones home; the network calls are yours
  • Local-first — code, sessions, and memory live on your disk as plain files
  • Any model — no favored vendor, no opaque routing, ever
  • You approve — writes, installs, pushes, and merges ask before they act