Route Claude CLI traffic through AI Gateway and Gemini

Create an AI Model Provider, and AI Model, and an AI Policy with all of your configuration

• The AI Model Provider entity defines your connection and stores your authentication credentials.
• The AI Request Transformer Policy removes extra fields that Gemini’s API does not support.
• The AI Model entity declares which upstream models are available, configures how client requests are routed, and specifies which AI Model Provider to use.

kongctl apply -f - --auto-approve --pat "$KONNECT_TOKEN" << 'EOF'
ai_gateway_model_providers:
  - ref: my-gemini-account
    ai_gateway: !lookup {id: !env AI_GATEWAY_ID}
    name: my-gemini-account
    display_name: "my-gemini-account"
    type: gemini
    config:
      auth:
        type: basic
        headers:
        - name: x-goog-api-key
          value: !secret {source: !env GEMINI_API_KEY}
ai_gateway_policies:
  - ref: strip-claude-beta-info
    ai_gateway: !lookup {id: !env AI_GATEWAY_ID}
    name: strip-claude-beta-info
    display_name: "strip-claude-beta-info"
    type: request-transformer-advanced
    config:
      remove:
        headers:
          - anthropic-beta
          - authorization
          - x-api-key
        querystring:
          - beta
        body:
          - output_config
          - context_management
          - mcp_servers
          - container
          - service_tier
          - reasoning_effort
ai_gateway_models:
  - ref: my-claude-gemini
    ai_gateway: !lookup {id: !env AI_GATEWAY_ID}
    name: my-claude-gemini
    display_name: "my-claude-gemini"
    type: model
    formats:
      - type: anthropic
    config:
      route:
        paths:
          - /
        model:
          body_param: model
          values:
            - my-claude-gemini
    targets:
      - name: gemini-2.5-flash
        provider: !ref my-gemini-account#name
        config:
          type: gemini
    policies:
      - !ref strip-claude-beta-info#name
    capabilities:
      - generate
EOF

The AI Model Provider uses the following settings:

• type: gemini: Specifies that this provider connects to the Gemini service using Gemini’s standard API format.
• name: my-gemini-account: A unique identifier that AI Models will reference to route requests through this provider.
• config.auth: Stores your Gemini API key. value: !secret {source: !env GEMINI_API_KEY} loads the value from your environment at apply time instead of embedding it in the YAML, and kongctl redacts it in plan and diff output. AI Gateway securely manages this credential and injects it into upstream requests automatically, eliminating the need for clients to pass API keys.

The AI Policy uses the following settings:

• type: request-transformer-advanced: Modifies requests before AI Gateway forwards them upstream.
• config.remove.headers: Removes the anthropic-beta, authorization, and x-api-key headers.
• config.remove.querystring: Removes the beta query string parameter.
• config.remove.body: Removes the output_config, context_management, mcp_servers, container, service_tier, and reasoning_effort body fields.

• Don’t strip model: AI Gateway uses that field to select the target, and removing it breaks routing.
• Claude Code beta features vary by version and may add other incompatible fields over time. If you still see an error mentioning an unexpected field after applying this Policy, add that field to the appropriate remove list and re-apply.

The AI Model uses the following settings:

• type: model: Specifies this is a synchronous model for request/response workloads.
• name: my-claude-gemini: A unique identifier for this model.
• formats: [type: anthropic]: Declares that this model accepts requests in Anthropic-compatible format, matching what Claude Code sends natively, even though the upstream model is Gemini.
• config.route.paths: [/]: Configures the custom base path where this model’s routes will be accessible. Setting this to a unique value avoids clashes when you have multiple AI Models.
• capabilities: [generate]: Enables the text generation capability. For a model using the anthropic format, the generate capability creates a /messages endpoint matching Anthropic’s native Messages API, so combined with your base path, clients send requests to /v1/messages.
• targets: Specifies which upstream AI Model Provider model to route requests to. Here, provider: !ref my-gemini-account#name references the AI Model Provider we created earlier, and name: gemini-2.5-flash specifies which Gemini model to call upstream.
• policies: [!ref strip-claude-beta-info#name]: Attaches the AI Policy created earlier so it applies to every request to this AI Model.

Claude Code’s experimental beta features send fields that Gemini rejects even with the AI Policy in place. Disable them, then start a session pointed at your local AI Gateway endpoint:

export CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS=1
export ANTHROPIC_BASE_URL=http://localhost:8000/

CLAUDE_CODE_DISABLE_UNKNOWN_MODEL_WINDOW_ENFORCEMENT=1 \
  claude --allowedTools "WebSearch,Read" --model "my-claude-gemini"

Tell me about the Madrid Skylitzes manuscript.

Claude Code might prompt you approve its web search for answering the question. When you select Yes, Claude will produce a full-length response to your request:

The Madrid Skylitzes is a remarkable 12th-century illuminated Byzantine
manuscript that represents one of the most important surviving examples
of medieval historical documentation. Here are the key details:

What it is

The Madrid Skylitzes is the only surviving illustrated manuscript of John
Skylitzes' "Synopsis of Histories" (Σύνοψις Ἱστοριῶν), which chronicles
Byzantine history from 811 to 1057 CE - covering the period from the death
of Emperor Nicephorus I to the deposition of Michael VI.

Artistic Significance

- 574 miniature paintings (with about 100 lost over time)
- Lavishly decorated with gold leaf, vibrant pigments, and intricate
detailing
- Depicts everything from imperial coronations and battles to daily life
in Byzantium
- The only surviving Byzantine illuminated chronicle written in Greek

Unique Collaboration

The manuscript is believed to be the work of 7 different artists from
various backgrounds:
- 4 Italian artists
- 1 English or French artist
- 2 Byzantine artists