diff --git a/services/site/src/docs/content/console/harnesses/claude-code.mdx b/services/site/src/docs/content/console/harnesses/claude-code.mdx index 3ba028577c7..1562e5849bb 100644 --- a/services/site/src/docs/content/console/harnesses/claude-code.mdx +++ b/services/site/src/docs/content/console/harnesses/claude-code.mdx @@ -7,6 +7,12 @@ Claude Code speaks the Anthropic Messages API. Point it at Console with two envi model, including GPT, Gemini, and open models. Console translates the request for models that don't accept Anthropic Messages. + + The [Console CLI](/console/harnesses#console-cli) sets up Claude Code with one command, including the workspace's + policies, MCP servers, skills and instructions. Run `npx @opencode/console setup claude-code`. The steps below set it + up by hand. + + ```bash export ANTHROPIC_BASE_URL="https://opencode.ai/inference/anthropic" export ANTHROPIC_AUTH_TOKEN="oc_sk_..." diff --git a/services/site/src/docs/content/console/harnesses/codex.mdx b/services/site/src/docs/content/console/harnesses/codex.mdx index 2e04811395e..feb5679e1b8 100644 --- a/services/site/src/docs/content/console/harnesses/codex.mdx +++ b/services/site/src/docs/content/console/harnesses/codex.mdx @@ -6,6 +6,11 @@ description: "Use Console and Go models from Codex." Codex speaks the OpenAI Responses API. Point it at Console and use any paid or Go model, including Claude, Gemini, and open models. Console translates the request for models that don't accept Responses. + + The [Console CLI](/console/harnesses#console-cli) sets up Codex with one command, including the workspace's policies, + MCP servers, skills and instructions. Run `npx @opencode/console setup codex`. The steps below set it up by hand. + + ## Configure Add Console as a model provider in `~/.codex/config.toml` and make it the default. diff --git a/services/site/src/docs/content/console/harnesses/hermes.mdx b/services/site/src/docs/content/console/harnesses/hermes.mdx index 6d61ab132b0..8c3116eca6c 100644 --- a/services/site/src/docs/content/console/harnesses/hermes.mdx +++ b/services/site/src/docs/content/console/harnesses/hermes.mdx @@ -6,6 +6,11 @@ description: "Use Console and Go models from Hermes Agent." [Hermes Agent](https://hermes-agent.nousresearch.com) includes OpenCode providers for pay-as-you-go and Go. It picks the right API for each paid or Go model. Free-tier chat models require OpenCode and return HTTP `403` from Hermes. + + The [Console CLI](/console/harnesses#console-cli) sets up Hermes with one command, including the workspace's policies, + MCP servers, skills and instructions. Run `npx @opencode/console setup hermes`. The steps below set it up by hand. + + | Provider | Plan | Key variable | | -------------- | ------------- | ---------------------- | | `opencode-zen` | Pay-as-you-go | `OPENCODE_ZEN_API_KEY` | diff --git a/services/site/src/docs/content/console/harnesses/index.mdx b/services/site/src/docs/content/console/harnesses/index.mdx index 43767430d89..06c2871137c 100644 --- a/services/site/src/docs/content/console/harnesses/index.mdx +++ b/services/site/src/docs/content/console/harnesses/index.mdx @@ -17,6 +17,36 @@ paid or Go model, and usage is billed to your workspace like any other request. Free-tier chat models are only available from within OpenCode. Other harnesses receive HTTP `403` for those models, even with a Console API key. Use paid models or a Go subscription. +## Console CLI + +The [Console CLI](https://github.com/anomalyco/opencode-console-cli) connects a harness to your workspace with one +command. It needs Node.js 22 or later. + +| Harness | Command | +| ---------------- | ---------------------------------------------- | +| Claude Code | `npx @opencode/console setup claude-code` | +| Codex | `npx @opencode/console setup codex` | +| Hermes | `npx @opencode/console setup hermes` | +| Pi | `npx @opencode/console setup pi` | +| DeepSeek Harness | `npx @opencode/console setup deepseek-harness` | + +Your browser opens Console, where you approve the sign-in and pick the workspace. The harness then gets + +- the workspace's models, through Console +- the workspace's policies, checked before every tool call +- the workspace's MCP servers, skills and instructions + +The CLI keeps the harness in sync with the workspace and updates itself. + +To keep a harness's own sign-in and models, such as a Claude or ChatGPT subscription, add `--keep-provider`. The harness +still gets the workspace's policies, MCP servers, skills and instructions. It works with Claude Code, Codex and Hermes. + +```bash +npx @opencode/console setup codex --keep-provider +``` + +To set up a harness by hand, or one the CLI doesn't cover, use an API key and the base URLs below. + ## Get an API key Harnesses authenticate with a Console API key. The quickest option is a diff --git a/services/site/src/docs/content/console/harnesses/pi.mdx b/services/site/src/docs/content/console/harnesses/pi.mdx index dce88cffb7e..9ae051d8758 100644 --- a/services/site/src/docs/content/console/harnesses/pi.mdx +++ b/services/site/src/docs/content/console/harnesses/pi.mdx @@ -6,6 +6,11 @@ description: "Use Console and Go models from the Pi coding agent." [Pi](https://pi.dev) includes OpenCode providers for pay-as-you-go and Go. It picks the right API for each paid or Go model. Free-tier chat models require OpenCode and return HTTP `403` from Pi. + + The [Console CLI](/console/harnesses#console-cli) sets up Pi with one command, including the workspace's policies, MCP + servers, skills and instructions. Run `npx @opencode/console setup pi`. The steps below set it up by hand. + + | Provider | Plan | | ------------- | ------------- | | `opencode` | Pay-as-you-go |