How to connect Kimi to DeepSeek Harness

Kimi is one of the providers DSH ships with, so setup is shorter than for a gateway: DSH already knows the endpoint, the protocol and the model list. You only supply the key.

Before you start

You need a partially funded Kimi (Moonshot) account, an API key, and Node.js installed. If DSH itself is not running yet, do the install guide first.

Step 1 — create an API key

Sign in to Kimi (Moonshot) and create a key in its console. Copy it once — most providers never show it again.

Open Kimi (Moonshot) →

Step 2 — start the harness

npx @deepseek-ai/dsh web

Web UI comes up at http://127.0.0.1:3080.

Step 3 — Add model provider → pick moonshotai

In Settings → Models, choose Add model provider and leave the card on Third-party model provider. The provider list shows DSH's built-in ids; moonshotai is the one for Kimi.

Step 4 — paste the key

The installed catalog supplies the endpoint, the protocol and the model list, so there is no base URL to type. If you leave the field blank, DSH tries a saved credential and then the documented environment variables.

Step 5 — verify

Pick a model in the composer and send a message. A yellow hint under the key field means no credential variable was detected — that is a hint, not a failure, and it does not validate anything. Detection also does not test the connection.

Or: configure it in YAML

The Web UI writes the same document you can edit directly. The profile for a standard dsh web launch is web, so the file is $DSH_HOME/profiles/web/cordis.patch.yml. Adapters re-read it on the next request — no restart needed.

# $DSH_HOME/profiles/web/cordis.patch.yml
# Kimi is a catalog provider, so only a key reference is needed.
# The key itself lives in $DSH_HOME/.credentials.yaml — never inline it here.
- id: llm-pi-ai
  config:
    providers:
      moonshotai:
        apiKeyEnv: MOONSHOT_API_KEY
        models:
          - id: kimi-k2-0905-preview
            input: [text]

apiKeyEnv names an environment variable rather than storing a secret in YAML. Export it before launching DSH.

Export the variable referenced above in the shell that starts DSH:

export DSH_API_KEY="sk-..."

Troubleshooting

• 404 on every request — the base URL is wrong. It must end at the version segment (/v1), never at /chat/completions.
• 401 Unauthorized — key revoked, or pasted with trailing whitespace. Note that DSH's env-var detection does not validate the key.
• CLIENT_UPGRADE_REQUIRED — the model needs a newer harness. The error names the version to install.
• Model missing from the picker — a saved default pointing at a deleted provider makes the composer show Select model and block input until you pick another.
• Empty or truncated reply — usually the model id resolved to a different model than you expected. Add the id explicitly rather than relying on Fetch available models.

Cost and caveats

Per token, same order of magnitude as DeepSeek. Long-context models cost more per token.

• Because it is built-in, you do not type a base URL — DSH supplies it from the installed catalog.
• Moonshot runs separate China (api.moonshot.cn) and international (api.moonshot.ai) platforms; the key must match the platform.
• Model names change; check the console before pinning one in config.

Plugins that pair well with this provider

Other providers

Source for every field above: the official DSH provider docs. If DSH changes a field name, that document wins over this page.