Add some docs fixes

Author: Shashikant86

docs/getting-started/installation.md

@@ -28,7 +28,7 @@ brew install uv
 ```
 
 !!! tip "Why uv?"
-    `uv` is 10-100x faster than pip for dependency resolution. `uv tool install` creates an isolated environment for CLI tools — no virtualenv management needed. If you don't have Python 3.11+ installed, uv can install it for you:
+    `uv` is 10-100x faster than pip for dependency resolution. `uv tool install` creates an isolated environment for CLI tools - no virtualenv management needed. If you don't have Python 3.11+ installed, uv can install it for you:
 
     ```bash
     uv python install 3.12
@@ -44,7 +44,7 @@ brew install uv
     uv tool install "rlm-code[tui,llm-all]"
     ```
 
-    This installs `rlm-code` as a globally available command in its own isolated environment. No virtualenv activation needed — just run `rlm-code` from anywhere.
+    This installs `rlm-code` as a globally available command in its own isolated environment. No virtualenv activation needed - just run `rlm-code` from anywhere.
 
 === "uv pip install"
 

docs/getting-started/quickstart.md

@@ -1,6 +1,6 @@
 # Quick Start
 
-This guide walks you through launching RLM Code, connecting to an LLM, running your first benchmark, validating results, and exploring the Research tab — all in under 10 minutes.
+This guide walks you through launching RLM Code, connecting to an LLM, running your first benchmark, validating results, and exploring the Research tab - all in under 10 minutes.
 
 ---
 

docs/index.md

@@ -8,7 +8,7 @@
 
 <span class="rlm-badge rlm-badge--purple">v0.1.5</span>
 <span class="rlm-badge rlm-badge--green">Python 3.11+</span>
-<span class="rlm-badge rlm-badge--blue">Apache-2.0</span>
+<span class="rlm-badge rlm-badge--blue">Apache 2.0</span>
 
 </div>
 
@@ -101,39 +101,39 @@ Risk assessment with **40+ rules**, 6 approval modes, and full audit logging to
 
 <div class="rlm-quickstart" markdown>
 
-### Install & Launch
+**Install and launch**
 
 ```bash
 uv tool install "rlm-code[tui,llm-all]"
 rlm-code
 ```
 
-### Connect to a model
+**Connect to a model**
 
-```
+```text
 /connect anthropic claude-opus-4-6
 ```
 
-### Run your first benchmark
+**Run your first benchmark**
 
-```
+```text
 /rlm bench preset=dspy_quick
 ```
 
-### Keep runs bounded
+**Keep runs bounded**
 
-```
+```text
 /rlm run "small scoped task" steps=4 timeout=30 budget=60
 /rlm abort all
 ```
 
-### Compare benchmark output
+**Compare benchmark output**
 
-```
+```text
 /rlm bench compare candidate=latest baseline=previous
 ```
 
-### Switch to the Research tab
+**Switch to the Research tab**
 
 Press `Ctrl+5` or `F6` to open the **Research** tab to see your run's dashboard, trajectory, reward curves, and live events.
 
@@ -181,27 +181,27 @@ graph TB
 
 ## 📋 Feature Matrix
 
-| Feature | Status | Module |
-|---------|--------|--------|
-| 🧠 RLM Runner (multi-paradigm) | ✅ | `rlm_code.rlm.runner` |
-| 🧪 Pure RLM Environment | ✅ | `rlm_code.rlm.pure_rlm_environment` |
-| 📡 Event System (27+ types) | ✅ | `rlm_code.rlm.events` |
-| 🎯 Policy Lab (16 policies) | ✅ | `rlm_code.rlm.policies` |
-| 🔒 HITL Approval Gates | ✅ | `rlm_code.rlm.approval` |
-| 📊 Observability (7 sinks) | ✅ | `rlm_code.rlm.observability` |
-| 🏆 Benchmarks (10 presets) | ✅ | `rlm_code.rlm.benchmarks` |
-| 📊 Leaderboard | ✅ | `rlm_code.rlm.leaderboard` |
-| ⏪ Session Replay | ✅ | `rlm_code.rlm.session_replay` |
-| 🔁 Paradigm Comparison | ✅ | `rlm_code.rlm.comparison` |
-| 📈 Trajectory Logging | ✅ | `rlm_code.rlm.trajectory` |
-| 🧹 Memory Compaction | ✅ | `rlm_code.rlm.memory_compaction` |
-| 📦 6 Sandbox Runtimes | ✅ | `rlm_code.sandbox.runtimes` |
-| 🤖 12+ LLM Providers | ✅ | `rlm_code.models` |
-| 🔌 MCP Server | ✅ | `rlm_code.mcp` |
-| 🖥️ Unified TUI (5 tabs) | ✅ | `rlm_code.ui.tui_app` |
-| ⌨️ 50+ Slash Commands | ✅ | `rlm_code.commands` |
-| ✅ Code Validation | ✅ | `rlm_code.validation` |
-| 🧩 Framework Adapters | ✅ | `rlm_code.rlm.frameworks` |
+| Feature | Module |
+|---------|--------|
+| 🧠 RLM Runner (multi-paradigm) | `rlm_code.rlm.runner` |
+| 🧪 Pure RLM Environment | `rlm_code.rlm.pure_rlm_environment` |
+| 📡 Event System (27+ types) | `rlm_code.rlm.events` |
+| 🎯 Policy Lab (16 policies) | `rlm_code.rlm.policies` |
+| 🔒 HITL Approval Gates | `rlm_code.rlm.approval` |
+| 📊 Observability (7 sinks) | `rlm_code.rlm.observability` |
+| 🏆 Benchmarks (10 presets) | `rlm_code.rlm.benchmarks` |
+| 📊 Leaderboard | `rlm_code.rlm.leaderboard` |
+| ⏪ Session Replay | `rlm_code.rlm.session_replay` |
+| 🔁 Paradigm Comparison | `rlm_code.rlm.comparison` |
+| 📈 Trajectory Logging | `rlm_code.rlm.trajectory` |
+| 🧹 Memory Compaction | `rlm_code.rlm.memory_compaction` |
+| 📦 6 Sandbox Runtimes | `rlm_code.sandbox.runtimes` |
+| 🤖 12+ LLM Providers | `rlm_code.models` |
+| 🔌 MCP Server | `rlm_code.mcp` |
+| 🖥️ Unified TUI (5 tabs) | `rlm_code.ui.tui_app` |
+| ⌨️ 50+ Slash Commands | `rlm_code.commands` |
+| Code Validation | `rlm_code.validation` |
+| 🧩 Framework Adapters | `rlm_code.rlm.frameworks` |
 
 ---
 

docs/stylesheets/extra.css

@@ -167,7 +167,7 @@
 }
 
 .rlm-hero-logo {
-  width: 180px;
+  width: min(340px, 72vw);
   height: auto;
   margin-bottom: 0.5em;
   filter: drop-shadow(0 0 24px rgba(124, 58, 237, 0.35));
@@ -243,11 +243,6 @@
   background: #050510;
 }
 
-.rlm-quickstart h3 {
-  margin-top: 0;
-  color: #a855f7;
-}
-
 /* Badge pills */
 .rlm-badge {
   display: inline-block;

tests/rlm/test_phase3.py

@@ -370,9 +370,9 @@ def test_benchmark_cases_have_pure_rlm_env(self):
         for preset in ["oolong_style", "browsecomp_style", "token_efficiency"]:
             cases = get_benchmark_cases(preset)
             for case in cases:
-                assert case.environment == "pure_rlm", (
-                    f"{preset}/{case.case_id} should use pure_rlm"
-                )
+                assert (
+                    case.environment == "pure_rlm"
+                ), f"{preset}/{case.case_id} should use pure_rlm"
 
     def test_benchmark_cases_have_reasonable_timeouts(self):
         """Test that paper benchmarks have appropriate timeouts."""

tests/test_retry.py

@@ -83,9 +83,9 @@ def failing_operation():
         with pytest.raises(ValueError):
             controller.execute_sync_with_retry(failing_operation, "test_op")
 
-        assert attempt_count == max_attempts, (
-            f"Expected {max_attempts} attempts, got {attempt_count}"
-        )
+        assert (
+            attempt_count == max_attempts
+        ), f"Expected {max_attempts} attempts, got {attempt_count}"
 
     @pytest.mark.asyncio
     async def test_async_retry_respects_max_attempts(self):

tests/test_security_validator.py

@@ -209,6 +209,6 @@ class MockConfig:
         validator = SecurityValidator()
 
         # Check module code (where calculator_tool is defined)
-        assert not validator.has_eval(program.module_code), (
-            f"Generated module code should not contain eval():\n{program.module_code}"
-        )
+        assert not validator.has_eval(
+            program.module_code
+        ), f"Generated module code should not contain eval():\n{program.module_code}"