@zenithfoundry/slm-gate 1.2.1

package/configs/cline-continue-opencode/README.md

@@ -0,0 +1,34 @@
+# Cline / Continue / Opencode MCP Configuration
+
+**File Location:** `cline_mcp.json` or `.continue/config.json` (depends on client, usually at the global or project root)
+
+```json
+{
+  "mcpServers": {
+    "slm-gate": {
+      "command": "node",
+      "args": [
+        "<ABS_PATH>/dist/mcp-gate/index.js"
+      ],
+      "env": {
+        "TLS_ADAPTER": "on",
+        "DOWNSTREAM_MCP": "{\"command\":\"node\",\"args\":[\"<ABS_PATH_TO_TLS>/dist/mcp-server.mjs\"]}"
+      }
+    }
+  }
+}
+```
+
+After adding this, make sure to build TLS first by running `pnpm run mcp:build` in your TLS directory, then restart/refresh MCP servers and verify with `slm-gate doctor`.
+
+The `env` values above apply to this MCP server only. The model gate below is shared by all your coding tools and reads its settings only from `slm-gate`'s own `.env`.
+
+## Model gate (Layer 2): send model requests through slm-gate
+
+Works with your own API key. Models bundled in the tool's own subscription can't be redirected. The model gate starts by itself when the tool starts the MCP server above.
+
+- **Cline / Roo Code:** Settings → API Provider **Anthropic** → tick **Use custom base URL** → `http://localhost:8787`
+- **Continue** (`~/.continue/config.yaml`, on the model): `provider: anthropic`, `apiBase: http://localhost:8787/v1/` (OpenAI models: `provider: openai`, `apiBase: http://localhost:8787/v1`)
+- **OpenCode** (`opencode.json`): `{ "provider": { "anthropic": { "options": { "baseURL": "http://localhost:8787/v1" } } } }`
+
+Changed `LLM_GATE_PORT`? `slm-gate doctor` prints these lines with the new port, and the lines for Kilo Code, Zed, Copilot's Custom Endpoint, Junie CLI and Aider.