rlm-code 0.1.2__tar.gz → 0.1.6__tar.gz

{rlm_code-0.1.2 → rlm_code-0.1.6}/CHANGELOG.md

@@ -5,7 +5,7 @@ All notable changes to this project are documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## [0.1.2] - 2026-02-20
+## [0.1.6] - 2026-02-20
 
 ### Added
 - Harness strategy selector with `tool_call` (default) and opt-in `codemode`.
@@ -29,7 +29,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Added explicit CodeMode guardrail policy documentation with blocked API classes and runtime limit defaults.
 - Codemode path remains opt-in; default harness behavior remains strict baseline `strategy=tool_call`.
 
-## [0.1.1] - 2026-02-15
+## [0.1.5] - 2026-02-15
 
 Initial public release of **RLM Code**.
 
@@ -55,4 +55,4 @@ Initial public release of **RLM Code**.
 - Unsafe local `exec` usage preserved only as an explicit, opt-in path for advanced development scenarios.
 
 [0.1.5]: https://github.com/SuperagenticAI/rlm-code/releases/tag/v0.1.5
-[0.1.2]: https://github.com/SuperagenticAI/rlm-code/releases/tag/v0.1.2
+[0.1.6]: https://github.com/SuperagenticAI/rlm-code/releases/tag/v0.1.6

{rlm_code-0.1.2 → rlm_code-0.1.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rlm-code
-Version: 0.1.2
+Version: 0.1.6
 Summary: RLM Code: Research Playground & Evaluation OS for Recursive Language Model Agentic Systems
 Project-URL: Homepage, https://github.com/SuperagenticAI/rlm-code
 Project-URL: Documentation, https://superagenticai.github.io/rlm-code/
@@ -118,7 +118,7 @@ RLM Code implements the [Recursive Language Models](https://arxiv.org/abs/2502.0
 
 RLM Code wraps this algorithm in an interactive terminal UI with built-in benchmarks, trajectory replay, and observability.
 
-## Release v0.1.2
+## Release v0.1.6
 
 This release adds the new CodeMode path as an opt-in harness strategy.
 

{rlm_code-0.1.2 → rlm_code-0.1.6}/README.md

@@ -25,7 +25,7 @@ RLM Code implements the [Recursive Language Models](https://arxiv.org/abs/2502.0
 
 RLM Code wraps this algorithm in an interactive terminal UI with built-in benchmarks, trajectory replay, and observability.
 
-## Release v0.1.2
+## Release v0.1.6
 
 This release adds the new CodeMode path as an opt-in harness strategy.
 

{rlm_code-0.1.2 → rlm_code-0.1.6}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "rlm-code"
-version = "0.1.2"
+version = "0.1.6"
 description = "RLM Code: Research Playground & Evaluation OS for Recursive Language Model Agentic Systems"
 readme = "README.md"
 license = "Apache-2.0"

{rlm_code-0.1.2 → rlm_code-0.1.6}/rlm_code/__init__.py

@@ -5,5 +5,5 @@ This package provides tools for creating, managing, and optimizing DSPy componen
 through natural language interactions.
 """
 
-__version__ = "0.1.2"
+__version__ = "0.1.6"
 __author__ = "Super Agentic AI"

{rlm_code-0.1.2 → rlm_code-0.1.6}/rlm_code/commands/slash_commands.py

@@ -1600,9 +1600,7 @@ class SlashCommandHandler:
                 show_warning_message("strategy=codemode requires mcp=on. Enabling MCP.")
                 include_mcp = True
             if strategy == "codemode" and allowlist:
-                show_warning_message(
-                    "tools=... allowlist is ignored for strategy=codemode."
-                )
+                show_warning_message("tools=... allowlist is ignored for strategy=codemode.")
                 allowlist = None
 
             console.print()

{rlm_code-0.1.2 → rlm_code-0.1.6}/rlm_code/harness/registry.py

@@ -333,7 +333,10 @@ class HarnessToolRegistry:
         normalized_server = server_name.strip().lower()
         normalized_tool = tool_name.strip().lower()
 
-        if self._mcp_allowed_servers is not None and normalized_server not in self._mcp_allowed_servers:
+        if (
+            self._mcp_allowed_servers is not None
+            and normalized_server not in self._mcp_allowed_servers
+        ):
             return (
                 False,
                 f"MCP tool '{full_name}' blocked by MCP policy (server '{server_name}' not allowed).",
@@ -351,11 +354,7 @@ class HarnessToolRegistry:
         if include_mcp and self.mcp_manager is not None:
             mcp_specs = self._list_mcp_specs()
             if self._mcp_allowed_tools is not None or self._mcp_allowed_servers is not None:
-                mcp_specs = [
-                    row
-                    for row in mcp_specs
-                    if self._is_mcp_tool_allowed(row.name)[0]
-                ]
+                mcp_specs = [row for row in mcp_specs if self._is_mcp_tool_allowed(row.name)[0]]
             specs = list(specs_map.values())
             specs.extend(mcp_specs)
             alias_candidates: dict[str, list[HarnessToolSpec]] = {}
@@ -985,7 +984,9 @@ def _domain_allowed(*, domain: str, allowed_domains: set[str], blocked_domains:
         return False
     if blocked_domains and any(domain == d or domain.endswith(f".{d}") for d in blocked_domains):
         return False
-    if allowed_domains and not any(domain == d or domain.endswith(f".{d}") for d in allowed_domains):
+    if allowed_domains and not any(
+        domain == d or domain.endswith(f".{d}") for d in allowed_domains
+    ):
         return False
     return True
 

{rlm_code-0.1.2 → rlm_code-0.1.6}/rlm_code/harness/runner.py

@@ -250,7 +250,9 @@ class HarnessRunner:
             mcp_tool_allowlist=mcp_tool_allowlist,
             mcp_server=mcp_server,
         )
-        server_name = self._resolve_codemode_server(tool_rows=tool_rows, requested_server=mcp_server)
+        server_name = self._resolve_codemode_server(
+            tool_rows=tool_rows, requested_server=mcp_server
+        )
         if not server_name:
             return HarnessRunResult(
                 completed=False,

{rlm_code-0.1.2 → rlm_code-0.1.6}/rlm_code/mcp/__init__.py

@@ -17,7 +17,7 @@ from .exceptions import (
 )
 from .session_wrapper import MCPSessionWrapper
 
-__version__ = "0.1.2"
+__version__ = "0.1.6"
 
 __all__ = [
     "MCPClientManager",

{rlm_code-0.1.2 → rlm_code-0.1.6}/rlm_code/rlm/benchmark_manager.py

@@ -174,7 +174,11 @@ class BenchmarkManagerMixin:
         """Execute a benchmark preset and persist aggregate summary."""
         resolved_mode = self._normalize_benchmark_mode(mode)
         resolved_harness_strategy = self._normalize_harness_strategy(harness_strategy)
-        if resolved_mode == "harness" and resolved_harness_strategy == "codemode" and not include_mcp:
+        if (
+            resolved_mode == "harness"
+            and resolved_harness_strategy == "codemode"
+            and not include_mcp
+        ):
             logger.warning("Harness codemode strategy requires MCP; enabling include_mcp.")
             include_mcp = True
         benchmark_id = datetime.now(timezone.utc).strftime("bench_%Y%m%d_%H%M%S_%f")
@@ -291,11 +295,7 @@ class BenchmarkManagerMixin:
             "mode": resolved_mode,
             "mcp_enabled": bool(include_mcp) if resolved_mode == "harness" else False,
             "mcp_server": str(mcp_server) if (resolved_mode == "harness" and mcp_server) else None,
-            "harness_strategy": (
-                resolved_harness_strategy
-                if resolved_mode == "harness"
-                else None
-            ),
+            "harness_strategy": (resolved_harness_strategy if resolved_mode == "harness" else None),
             "source": extra_sources.get(str(preset).strip().lower(), "builtin"),
             "description": extra_descriptions.get(str(preset).strip().lower(), ""),
             "pack_paths": [str(item) for item in (pack_paths or self._benchmark_pack_paths)],
@@ -961,12 +961,11 @@ class BenchmarkManagerMixin:
         for step in tool_steps:
             if step.tool_result is None:
                 continue
-            metadata = step.tool_result.metadata if isinstance(step.tool_result.metadata, dict) else {}
+            metadata = (
+                step.tool_result.metadata if isinstance(step.tool_result.metadata, dict) else {}
+            )
             resolved_name = str(
-                metadata.get("tool_full_name")
-                or metadata.get("resolved_tool")
-                or step.tool
-                or ""
+                metadata.get("tool_full_name") or metadata.get("resolved_tool") or step.tool or ""
             ).strip()
             if resolved_name.startswith("mcp:"):
                 mcp_tool_calls += 1
@@ -982,8 +981,7 @@ class BenchmarkManagerMixin:
             ):
                 codemode_discovery_calls += 1
         codemode_guardrail_blocked = any(
-            str(step.action) == "codemode_plan"
-            and "guardrail" in str(step.reasoning or "").lower()
+            str(step.action) == "codemode_plan" and "guardrail" in str(step.reasoning or "").lower()
             for step in result.steps
         )
         return {

{rlm_code-0.1.2 → rlm_code-0.1.6}/tests/test_harness_runner.py

@@ -156,7 +156,9 @@ def test_harness_run_supports_codemode_strategy(tmp_path: Path) -> None:
     manager = _CodeModeMCPManager()
     runner = HarnessRunner(llm_connector=connector, mcp_manager=manager, workdir=tmp_path)
 
-    result = runner.run(task="Get weather in SF", strategy="codemode", include_mcp=True, max_steps=4)
+    result = runner.run(
+        task="Get weather in SF", strategy="codemode", include_mcp=True, max_steps=4
+    )
 
     assert result.completed is True
     assert "San Francisco" in result.final_response

{rlm_code-0.1.2 → rlm_code-0.1.6}/tests/test_provider_registry.py

@@ -186,9 +186,14 @@ def test_connector_supported_provider_metadata_includes_connection_type(tmp_path
     assert isinstance(openai["example_models"], list)
 
 
-def test_connector_uses_superqode_model_catalog(tmp_path: Path):
+def test_connector_uses_superqode_model_catalog(tmp_path: Path, monkeypatch):
     """Provider model list should align with SuperQode catalog where available."""
     connector = make_connector(tmp_path)
+    monkeypatch.setattr(
+        connector,
+        "_list_opencode_models_cached",
+        lambda timeout=0.6: [],
+    )
 
     openai_models = connector.list_provider_example_models("openai", limit=3)
     gemini_models = connector.list_provider_example_models("gemini", limit=2)