Aider

Aider is a command-line pair-programming tool that edits code in your local git repo. It reaches OpenAI-compatible endpoints through LiteLLM, so it's a first-class connect target: the proxy writes the config for you.

Before you start

The proxy must be running (zs-proxy proxy start) and funded — see How-to guides.

The quick way: zs-proxy connect aider

zs-proxy connect aider                          # configure aider to use the proxy
zs-proxy connect aider --model qwen3-coder      # ...and set it as the default model
zs-proxy connect aider --print                  # preview the merged config, write nothing

This merges the proxy endpoint into your existing ~/.aider.conf.yml (backing the original up to .aider.conf.yml.bak). Your other aider settings are preserved. What it writes:

openai-api-base: http://127.0.0.1:9376/v1
openai-api-key: zerosignal-local     # placeholder — the proxy ignores it
model-metadata-file: ~/.aider.model.metadata.json   # written as an absolute path
model: openai/qwen3-coder            # only when you pass --model

It also writes a second file next to that one, ~/.aider.model.metadata.json, and points model-metadata-file at it. Aider reaches models through LiteLLM, whose catalog knows nothing about the network's models — without the sidecar it warns "Unknown context window size and costs" and falls back to generic behavior. The sidecar carries each model's context window and costs under the same openai/<id> name. Entries are upserted, so any you wrote by hand for other models survive, and --print renders both files.

info

Aider routes through LiteLLM, so there's no named provider block — it selects the OpenAI-compatible path from the openai-api-base / openai-api-key keys, and model ids carry an openai/ prefix (e.g. openai/qwen3-coder). Pick any id from the live catalog and prefix it that way.

Select the model and run

Start aider in a git repo. Pass a model with --model or switch in-session:

aider --model openai/qwen3-coder
# or, once running:
# /model openai/qwen3-coder

Ask for a change; a reply confirms the setup works.

Troubleshooting

Symptom	Fix
connect says aider wasn't detected	Name it explicitly — zs-proxy connect aider configures the named tool even without detection, writing ~/.aider.conf.yml.
"model not found" / provider errors	Use the openai/ prefix — aider needs openai/<id>, not a bare <id>. List ids with curl http://localhost:9376/v1/models.
Connection refused / can't reach the endpoint	The proxy isn't running or the base URL is wrong. Confirm with curl http://localhost:9376/healthz and start it with zs-proxy proxy start.
wallet_unfunded / payment errors	Add funds in the chat app, or with zs-proxy fund for a proxy-only wallet — see Wallet & funding.

What's next

• Connecting AI tools — every connect flag and the other supported tools.
• Routing preferences — pin an operator or set a price ceiling.
• Pricing — what an agent turn costs.
• How-to guides — connect another app.