refactor(cli): remove stale cli provider leftovers

Author: steipete

docs/concepts/context.md

@@ -167,7 +167,7 @@ pluggable interface, lifecycle hooks, and configuration.
 `/context` prefers the latest **run-built** system prompt report when available:
 
 - `System prompt (run)` = captured from the last embedded (tool-capable) run and persisted in the session store.
-- `System prompt (estimate)` = computed on the fly when no run report exists (or when running via a CLI backend that doesn’t generate the report).
+- `System prompt (estimate)` = computed on the fly when no run report exists yet.
 
 Either way, it reports sizes and top contributors; it does **not** dump the full system prompt or tool schemas.
 

docs/gateway/configuration-reference.md

@@ -1064,7 +1064,6 @@ Z.AI GLM-4.x models automatically enable thinking mode unless you set `--thinkin
 Z.AI models enable `tool_stream` by default for tool call streaming. Set `agents.defaults.models["zai/<model>"].params.tool_stream` to `false` to disable it.
 Anthropic Claude 4.6 models default to `adaptive` thinking when no explicit thinking level is set.
 
-- CLI backends are text-first; tools are always disabled.
 - Sessions supported when `sessionArg` is set.
 - Image pass-through supported when `imageArg` accepts file paths.
 

docs/plugins/architecture.md

@@ -30,7 +30,6 @@ native OpenClaw plugin registers against one or more capability types:
 | Capability             | Registration method                              | Example plugins                      |
 | ---------------------- | ------------------------------------------------ | ------------------------------------ |
 | Text inference         | `api.registerProvider(...)`                      | `openai`, `anthropic`                |
-| CLI inference backend  | `api.registerCliBackend(...)`                    | `openai`, `anthropic`                |
 | Speech                 | `api.registerSpeechProvider(...)`                | `elevenlabs`, `microsoft`            |
 | Realtime transcription | `api.registerRealtimeTranscriptionProvider(...)` | `openai`                             |
 | Realtime voice         | `api.registerRealtimeVoiceProvider(...)`         | `openai`                             |

docs/plugins/sdk-overview.md

@@ -122,7 +122,6 @@ explicitly promotes one as public.
     | `plugin-sdk/provider-entry` | `defineSingleProviderPluginEntry` |
     | `plugin-sdk/provider-setup` | Curated local/self-hosted provider setup helpers |
     | `plugin-sdk/self-hosted-provider-setup` | Focused OpenAI-compatible self-hosted provider setup helpers |
-    | `plugin-sdk/cli-backend` | CLI backend defaults + watchdog constants |
     | `plugin-sdk/provider-auth-runtime` | Runtime API-key resolution helpers for provider plugins |
     | `plugin-sdk/provider-auth-api-key` | API-key onboarding/profile-write helpers |
     | `plugin-sdk/provider-auth-result` | Standard OAuth auth-result builder |
@@ -283,7 +282,6 @@ methods:
 | Method                                           | What it registers                |
 | ------------------------------------------------ | -------------------------------- |
 | `api.registerProvider(...)`                      | Text inference (LLM)             |
-| `api.registerCliBackend(...)`                    | Local CLI inference backend      |
 | `api.registerChannel(...)`                       | Messaging channel                |
 | `api.registerSpeechProvider(...)`                | Text-to-speech / STT synthesis   |
 | `api.registerRealtimeTranscriptionProvider(...)` | Streaming realtime transcription |

docs/reference/session-management-compaction.md

@@ -332,7 +332,7 @@ Notes:
 - The default prompt/system prompt include a `NO_REPLY` hint to suppress
   delivery.
 - The flush runs once per compaction cycle (tracked in `sessions.json`).
-- The flush runs only for embedded Pi sessions (CLI backends skip it).
+- The flush runs only for embedded Pi sessions.
 - The flush is skipped when the session workspace is read-only (`workspaceAccess: "ro"` or `"none"`).
 - See [Memory](/concepts/memory) for the workspace file layout and write patterns.
 

docs/tools/acp-agents.md

@@ -114,7 +114,6 @@ Important distinction:
   For operators, the practical rule is:
 
 - want `/acp spawn`, bindable sessions, runtime controls, or persistent harness work: use ACP
-- want simple local text fallback through the raw CLI: use CLI backends
 
 ## Bound sessions
 

scripts/audit-seams.mjs

@@ -583,7 +583,6 @@ function describeCronSeamKinds(relativePath, source) {
 
   const seamKinds = [];
   const importsAgentRunner = hasAnyImportSource(source, [
-    "../../agents/cli-runner.js",
     "../../agents/pi-embedded.js",
     "../../agents/model-fallback.js",
     "../../agents/subagent-registry.js",
@@ -625,9 +624,7 @@ function describeCronSeamKinds(relativePath, source) {
 
   if (
     importsAgentRunner &&
-    /\brunCliAgent\b|\brunEmbeddedPiAgent\b|\brunWithModelFallback\b|\bregisterAgentRunContext\b/.test(
-      source,
-    )
+    /\brunEmbeddedPiAgent\b|\brunWithModelFallback\b|\bregisterAgentRunContext\b/.test(source)
   ) {
     seamKinds.push("cron-agent-handoff");
   }

scripts/e2e/plugins-docker.sh

@@ -908,8 +908,6 @@ if (!inspect.gatewayMethods.includes("demo.marketplace.shortcut.v2")) {
 console.log("ok");
 NODE
 
-echo "Running bundle MCP CLI-agent e2e..."
-pnpm exec vitest run --config vitest.e2e.config.ts src/agents/cli-runner.bundle-mcp.e2e.test.ts
 EOF
 
 echo "OK"

src/agents/command/session-store.test.ts

@@ -1,10 +1,8 @@
 import { describe, it, expect, vi } from "vitest";
-import type { OpenClawConfig } from "../config/config.js";
 import { resetLogger, setLoggerOverride } from "../logging/logger.js";
 import {
   buildAllowedModelSet,
   inferUniqueProviderFromConfiguredModels,
-  isCliProvider,
   parseModelRef,
   buildModelAliasIndex,
   normalizeModelSelection,
@@ -101,7 +99,7 @@ function createProviderWithModelsConfig(provider: string, models: Array<Record<s
 
 function resolveConfiguredRefForTest(cfg: Partial<OpenClawConfig>) {
   return resolveConfiguredModelRef({
-    cfg: cfg as OpenClawConfig,
+    cfg: cfg,
     defaultProvider: "openai",
     defaultModel: "gpt-5.4",
   });
@@ -131,12 +129,6 @@ describe("model-selection", () => {
     });
   });
 
-  describe("isCliProvider", () => {
-    it("returns false for provider ids", () => {
-      expect(isCliProvider("example-cli", {} as OpenClawConfig)).toBe(false);
-    });
-  });
-
   describe("modelKey", () => {
     it("keeps canonical OpenRouter native ids without duplicating the provider", () => {
       expect(modelKey("openrouter", "openrouter/hunter-alpha")).toBe("openrouter/hunter-alpha");
@@ -462,7 +454,7 @@ describe("model-selection", () => {
       };
 
       const index = buildModelAliasIndex({
-        cfg: cfg as OpenClawConfig,
+        cfg: cfg,
         defaultProvider: "anthropic",
       });
 
@@ -751,7 +743,7 @@ describe("model-selection", () => {
         };
 
         const result = resolveConfiguredModelRef({
-          cfg: cfg as OpenClawConfig,
+          cfg: cfg,
           defaultProvider: "google",
           defaultModel: "gemini-pro",
         });
@@ -779,7 +771,7 @@ describe("model-selection", () => {
         };
 
         const result = resolveConfiguredModelRef({
-          cfg: cfg as OpenClawConfig,
+          cfg: cfg,
           defaultProvider: "google",
           defaultModel: "gemini-pro",
         });
@@ -832,7 +824,7 @@ describe("model-selection", () => {
     it("should use default provider/model if config is empty", () => {
       const cfg: Partial<OpenClawConfig> = {};
       const result = resolveConfiguredModelRef({
-        cfg: cfg as OpenClawConfig,
+        cfg: cfg,
         defaultProvider: "openai",
         defaultModel: "gpt-4",
       });
@@ -912,7 +904,7 @@ describe("model-selection", () => {
         };
 
         const result = resolveConfiguredModelRef({
-          cfg: cfg as OpenClawConfig,
+          cfg: cfg,
           defaultProvider: "openai",
           defaultModel: "gpt-5.4",
         });