PrimeIntellect-ai · willccbb · Jan 30, 2026 · Jan 29, 2026 · Jan 29, 2026
diff --git a/environments/README.md b/environments/README.md
@@ -44,7 +44,6 @@ This folder contains installable example environments that showcase common usage
   - **rlm_secrets**: Puzzle environment testing RLM functionality including root-level tools, sub-LLM tool use, and file operations.
 
 - **HarborEnv / CliAgentEnv (CLI agent sandboxes)**
-  - **dummy_harbor_env**: Minimal Harbor environment for testing the CLI agent interception framework.
   - **opencode_harbor**: Runs the OpenCode CLI agent on Harbor tasks with API interception via Prime Tunnel.
   - **terminus_harbor**: Runs the Terminus agent on Harbor tasks with API interception via Prime Tunnel.
 
@@ -74,7 +73,7 @@ This folder contains installable example environments that showcase common usage
 - **ToolEnv with real tools**: `wiki_search`, `math_python`
 - **Custom MultiTurnEnv**: `alphabet_sort`, `doublecheck`, `sentence_repeater`, `wordle`
 - **GymEnv integration**: `gem_wordle`
-- **CLI agent sandboxes**: `dummy_harbor_env`, `opencode_harbor`, `terminus_harbor`
+- **CLI agent sandboxes**: `opencode_harbor`, `terminus_harbor`
 - **MCP integration**: `mcp_search_env`
 - **RLM (recursive LLM)**: `rlm_secrets`
 - **Environment and rubric composition**: `math_group`, `math_python`, `wiki_search`

diff --git a/environments/dummy_harbor_env/README.md b/environments/dummy_harbor_env/README.md
diff --git a/environments/dummy_harbor_env/dummy_harbor_env.py b/environments/dummy_harbor_env/dummy_harbor_env.py
diff --git a/environments/dummy_harbor_env/pyproject.toml b/environments/dummy_harbor_env/pyproject.toml
diff --git a/environments/dummy_harbor_env/tasks/hello-world/environment/Dockerfile b/environments/dummy_harbor_env/tasks/hello-world/environment/Dockerfile
diff --git a/environments/dummy_harbor_env/tasks/hello-world/instruction.md b/environments/dummy_harbor_env/tasks/hello-world/instruction.md
diff --git a/environments/dummy_harbor_env/tasks/hello-world/solution/solve.sh b/environments/dummy_harbor_env/tasks/hello-world/solution/solve.sh
diff --git a/environments/dummy_harbor_env/tasks/hello-world/task.toml b/environments/dummy_harbor_env/tasks/hello-world/task.toml
diff --git a/environments/dummy_harbor_env/tasks/hello-world/tests/test.sh b/environments/dummy_harbor_env/tasks/hello-world/tests/test.sh
diff --git a/environments/dummy_harbor_env/tasks/hello-world/tests/test_state.py b/environments/dummy_harbor_env/tasks/hello-world/tests/test_state.py
diff --git a/verifiers/envs/experimental/cli_agent_env.py b/verifiers/envs/experimental/cli_agent_env.py
@@ -156,7 +156,7 @@ async def setup_state(self, state: State) -> State:
         state["sandbox_id"] = sandbox.id
         self.active_sandboxes.add(sandbox.id)
         logger.debug(f"Created sandbox {sandbox.id}")
-        await sandbox_client.wait_for_creation(sandbox.id)
+        await sandbox_client.wait_for_creation(sandbox.id, max_attempts=120)
 
         await self.post_sandbox_setup(state, sandbox_client)