Connect your tool

Ready-made configs and prompts for agents and IDEs: Claude Code, Codex CLI, Cursor, Cline and the rest.

Every tool below has a config, and most also have a prompt: paste it into the tool and its own agent edits its config. Substitute your own key; everything else in the samples is already ours.

If a tool is not listed, look at which request shape it speaks: an OpenAI-compatible client needs only base_url and a key; an Anthropic client needs the same against /v1/messages.

Configs, per tool

Gives your assistant tools instead of curl: it picks a model itself, generates and shows the picture. The server is remote, nothing to install — the URL and your key are enough. Video returns a job id the assistant polls on its own.

Option 1 · paste this prompt into the tool
Add an MCP server named "teamtoken" to my client config: type "http", url "https://teamtoken.store/mcp", header x-api-key = "sk-…" (my TeamToken key). Keep other servers intact.
Option 2 · edit the config by hand
// Claude Desktop / Claude Code / Cursor / Cline — одна и та же форма
// the same shape everywhere
{ "mcpServers": {
    "teamtoken": {
      "type": "http",
      "url": "https://teamtoken.store/mcp",
      "headers": { "x-api-key": "sk-…" }
    }
} }

// инструменты / tools: list_models, generate_image, edit_image,
//                      generate_video, extend_video, get_job, get_balance

Claude Code speaks the Anthropic-style /v1/messages — TeamToken accepts it. Any model from the catalog works. Restart after changing settings.

Option 1 · paste this prompt into the tool
Add an "env" block to my ~/.claude/settings.json so Claude Code targets a custom endpoint: ANTHROPIC_BASE_URL="https://api.teamtoken.store", ANTHROPIC_AUTH_TOKEN="sk-…" (my TeamToken API key), ANTHROPIC_MODEL="gpt-5.6-sol". Keep existing settings intact and tell me to restart Claude Code.
Option 2 · edit the config by hand
# вариант через окружение / via shell env
export ANTHROPIC_BASE_URL="https://api.teamtoken.store"
export ANTHROPIC_AUTH_TOKEN="sk-…"
export ANTHROPIC_MODEL="gpt-5.6-sol"
claude

# или / or ~/.claude/settings.json
{ "env": {
    "ANTHROPIC_BASE_URL": "https://api.teamtoken.store",
    "ANTHROPIC_AUTH_TOKEN": "sk-…",
    "ANTHROPIC_MODEL": "gpt-5.6-sol"
} }

Since 2026 Codex only supports wire_api = "responses" — TeamToken handles /v1/responses. The key must come from an env var.

Option 1 · paste this prompt into the tool
Update my ~/.codex/config.toml: add a model provider "teamtoken" with name "TeamToken", base_url "https://api.teamtoken.store/v1", env_key "TEAMTOKEN_KEY", wire_api "responses"; set top-level model = "gpt-5.6-sol" and model_provider = "teamtoken". Then remind me to export TEAMTOKEN_KEY with my key.
Option 2 · edit the config by hand
# ~/.codex/config.toml
model = "gpt-5.6-sol"
model_provider = "teamtoken"

[model_providers.teamtoken]
name = "TeamToken"
base_url = "https://api.teamtoken.store/v1"
env_key = "TEAMTOKEN_KEY"
wire_api = "responses"

# в окружении / in your shell:
export TEAMTOKEN_KEY="sk-…"

~/.config/opencode/opencode.json — a custom OpenAI-compatible provider; pick the model with /models.

Option 1 · paste this prompt into the tool
Add a provider "teamtoken" to my ~/.config/opencode/opencode.json: npm "@ai-sdk/openai-compatible", name "TeamToken", options.baseURL "https://api.teamtoken.store/v1", options.apiKey "{env:TEAMTOKEN_KEY}", models gpt-5.6-sol / gpt-5.4 / gemini-3.1-pro. Merge with my existing config, then remind me to export TEAMTOKEN_KEY.
Option 2 · edit the config by hand
{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "teamtoken": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "TeamToken",
      "options": {
        "baseURL": "https://api.teamtoken.store/v1",
        "apiKey": "{env:TEAMTOKEN_KEY}"
      },
      "models": { "gpt-5.6-sol": {}, "gpt-5.4": {}, "gemini-3.1-pro": {} }
    }
  }
}

Configured in the VS Code extension UI — there is no config file to prompt the agent about.

Setup
Settings → API Provider: "OpenAI Compatible"
  Base URL:  https://api.teamtoken.store/v1
  API Key:   sk-…  (ваш ключ / your key)
  Model ID:  gpt-5.6-sol

Cursor overrides the OpenAI base URL in its settings UI. Some built-in features (Tab, Apply) only work with Cursor-native models.

Setup
Cursor Settings → Models → API Keys
  OpenAI API Key:           sk-…  (ваш ключ / your key)
  Override OpenAI Base URL: https://api.teamtoken.store/v1
  + Add model: gpt-5.6-sol  (и выключите остальные / disable the rest)

~/.continue/config.yaml — the openai provider with a custom apiBase.

Option 1 · paste this prompt into the tool
Add a model to my ~/.continue/config.yaml: name "gpt-5.6-sol (TeamToken)", provider "openai", model "gpt-5.6-sol", apiBase "https://api.teamtoken.store/v1", apiKey "sk-…" (my TeamToken key). Keep the rest of the config as is.
Option 2 · edit the config by hand
# ~/.continue/config.yaml
models:
  - name: gpt-5.6-sol (TeamToken)
    provider: openai
    model: gpt-5.6-sol
    apiBase: https://api.teamtoken.store/v1
    apiKey: sk-…

Standard OpenAI env vars are enough; prefix the model with openai/.

Option 1 · paste this prompt into the tool
Add to my shell profile (~/.bashrc or ~/.zshrc): export OPENAI_API_BASE="https://api.teamtoken.store/v1" and export OPENAI_API_KEY="sk-…" (my TeamToken key), and create an alias to run aider with --model openai/gpt-5.6-sol.
Option 2 · edit the config by hand
export OPENAI_API_BASE="https://api.teamtoken.store/v1"
export OPENAI_API_KEY="sk-…"
aider --model openai/gpt-5.6-sol

Nous Hermes reads ~/.hermes/cli-config.yaml. Point its default model at a custom OpenAI-compatible provider; values in .env override the file. Restart Hermes after editing.

Option 1 · paste this prompt into the tool
Edit my ~/.hermes/cli-config.yaml: set model.default "gpt-5.6-sol", model.provider "custom", model.base_url "https://api.teamtoken.store/v1", model.api_key "sk-…" (my TeamToken key). Keep the rest and remind me to restart Hermes.
Option 2 · edit the config by hand
# ~/.hermes/cli-config.yaml
model:
  default: "gpt-5.6-sol"
  provider: "custom"
  base_url: "https://api.teamtoken.store/v1"
  api_key: "sk-…"

~/.openclaw/openclaw.json (JSON5): add a custom openai-completions provider, then select it with openclaw models set teamtoken/gpt-5.6-sol.

Option 1 · paste this prompt into the tool
Add a provider "teamtoken" to my ~/.openclaw/openclaw.json under models.providers: baseUrl "https://api.teamtoken.store/v1", apiKey "${TEAMTOKEN_KEY}", api "openai-completions", a model gpt-5.6-sol; set agents.defaults.model.primary to "teamtoken/gpt-5.6-sol". Merge with my existing config, then remind me to export TEAMTOKEN_KEY.
Option 2 · edit the config by hand
// ~/.openclaw/openclaw.json
{
  "models": {
    "mode": "merge",
    "providers": {
      "teamtoken": {
        "baseUrl": "https://api.teamtoken.store/v1",
        "apiKey": "${TEAMTOKEN_KEY}",
        "api": "openai-completions",
        "models": [{ "id": "gpt-5.6-sol", "name": "gpt-5.6-sol (TeamToken)" }]
      }
    }
  },
  "agents": { "defaults": { "model": { "primary": "teamtoken/gpt-5.6-sol" } } }
}

# в окружении / in your shell:
export TEAMTOKEN_KEY="sk-…"

Block's Goose works with any OpenAI-compatible API via its openai provider. OPENAI_HOST is the host only — Goose appends /v1/chat/completions — so use the origin WITHOUT /v1. Pick the model with GOOSE_MODEL. Restart Goose after changing settings.

Option 1 · paste this prompt into the tool
Configure Goose for a custom OpenAI-compatible gateway. Set env (and ~/.config/goose/config.yaml): GOOSE_PROVIDER="openai", GOOSE_MODEL="gpt-5.6-sol", OPENAI_HOST="https://api.teamtoken.store" (host without /v1), OPENAI_API_KEY="sk-…" (my TeamToken key). Keep my other settings and remind me to restart Goose.
Option 2 · edit the config by hand
# вариант через окружение / via shell env
export GOOSE_PROVIDER="openai"
export GOOSE_MODEL="gpt-5.6-sol"
export OPENAI_HOST="https://api.teamtoken.store"   # хост без /v1 / host only, no /v1 — Goose appends /v1/chat/completions itself
export OPENAI_API_KEY="sk-…"
goose

# или / or ~/.config/goose/config.yaml
GOOSE_PROVIDER: openai
GOOSE_MODEL: gpt-5.6-sol
OPENAI_HOST: https://api.teamtoken.store
OPENAI_API_KEY: sk-…