RLM Infra

Author: Shashikant86

.github/workflows/ci.yml

@@ -18,56 +18,37 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
-
       - name: Install uv
         uses: astral-sh/setup-uv@v4
         with:
           version: "latest"
 
       - name: Install dependencies
-        run: |
-          uv venv
-          uv pip install -e ".[dev]"
+        run: uv sync --locked --dev
 
       - name: Run ruff linter
-        run: |
-          source .venv/bin/activate
-          ruff check rlm_code tests
+        run: uv run ruff check rlm_code tests
 
       - name: Run ruff formatter check
-        run: |
-          source .venv/bin/activate
-          ruff format --check rlm_code tests
+        run: uv run ruff format --check rlm_code tests
 
   typecheck:
     name: Type Check
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
-
       - name: Install uv
         uses: astral-sh/setup-uv@v4
         with:
           version: "latest"
 
       - name: Install dependencies
-        run: |
-          uv venv
-          uv pip install -e ".[dev]"
+        run: uv sync --locked --dev
 
       - name: Run mypy on core modules
         run: |
-          source .venv/bin/activate
-          mypy rlm_code/core/config.py rlm_code/core/debug_logger.py rlm_code/mcp/utils.py rlm_code/mcp/retry.py rlm_code/models/cache.py rlm_code/models/streaming.py rlm_code/validation/security.py --ignore-missing-imports
+          uv run mypy rlm_code/core/config.py rlm_code/core/debug_logger.py rlm_code/mcp/utils.py rlm_code/mcp/retry.py rlm_code/models/cache.py rlm_code/models/streaming.py rlm_code/validation/security.py --ignore-missing-imports
 
   test:
     name: Test - Python ${{ matrix.python-version }} on ${{ matrix.os }}
@@ -76,30 +57,21 @@ jobs:
       fail-fast: false
       matrix:
         os: [ubuntu-latest, macos-latest]
-        python-version: ["3.10", "3.11", "3.12", "3.13"]
+        python-version: ["3.11", "3.12", "3.13"]
 
     steps:
       - uses: actions/checkout@v4
 
-      - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@v5
-        with:
-          python-version: ${{ matrix.python-version }}
-
       - name: Install uv
         uses: astral-sh/setup-uv@v4
         with:
           version: "latest"
 
       - name: Install dependencies
-        run: |
-          uv venv
-          uv pip install -e ".[test]"
+        run: uv sync --locked --python ${{ matrix.python-version }} --extra test
 
       - name: Run tests
-        run: |
-          source .venv/bin/activate || .venv\Scripts\activate
-          pytest tests/ -v --cov=rlm_code --cov-report=xml --cov-report=term-missing
+        run: uv run pytest tests/ -v --cov=rlm_code --cov-report=xml --cov-report=term-missing
 
       - name: Upload coverage
         if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.12'
@@ -115,25 +87,17 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
-
       - name: Install uv
         uses: astral-sh/setup-uv@v4
         with:
           version: "latest"
 
       - name: Install dependencies
-        run: |
-          uv venv
-          uv pip install -e ".[test]"
+        run: uv sync --locked --extra test
 
       - name: Run deterministic RLM benchmark gate
         run: |
-          source .venv/bin/activate
-          python scripts/rlm_bench_gate.py \
+          uv run python scripts/rlm_bench_gate.py \
             --baseline tests/fixtures/rlm_ci_baseline_generic_smoke.json \
             --preset generic_smoke \
             --limit 2
@@ -145,27 +109,17 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
-
       - name: Install uv
         uses: astral-sh/setup-uv@v4
         with:
           version: "latest"
 
-      - name: Install build dependencies
-        run: |
-          uv pip install --system build hatchling
-
       - name: Build package
-        run: python -m build
+        run: uv build
 
       - name: Check distribution
         run: |
-          uv pip install --system twine
-          twine check dist/*
+          uv tool run twine check dist/*
 
       - name: Upload build artifacts
         uses: actions/upload-artifact@v4

.github/workflows/deploy-docs.yml

@@ -24,23 +24,16 @@ jobs:
         with:
           fetch-depth: 0
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: '3.12'
-
       - name: Install uv
         uses: astral-sh/setup-uv@v4
         with:
           version: "latest"
 
-      - name: Install MkDocs and dependencies
-        run: |
-          uv pip install --system mkdocs-material mkdocs-minify-plugin
+      - name: Install dependencies
+        run: uv sync --locked --extra docs
 
       - name: Build and deploy documentation
-        run: |
-          mkdocs gh-deploy --force --clean --verbose
+        run: uv run mkdocs gh-deploy --force --clean --verbose
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 

.github/workflows/pre-commit.yml

@@ -12,18 +12,13 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
-
       - name: Install uv
         uses: astral-sh/setup-uv@v4
         with:
           version: "latest"
 
       - name: Install pre-commit
-        run: uv pip install --system pre-commit
+        run: uv tool install pre-commit
 
       - name: Cache pre-commit environments
         uses: actions/cache@v4

.github/workflows/release.yml

@@ -0,0 +1,51 @@
+name: Release
+
+on:
+  push:
+    tags:
+      - "v*"
+  workflow_dispatch:
+
+jobs:
+  build:
+    name: Build Distributions
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v4
+        with:
+          version: "latest"
+
+      - name: Build package
+        run: uv build
+
+      - name: Check distributions
+        run: uv tool run twine check dist/*
+
+      - name: Upload distributions
+        uses: actions/upload-artifact@v4
+        with:
+          name: release-dist
+          path: dist/
+          retention-days: 7
+
+  publish:
+    name: Publish to PyPI
+    needs: build
+    runs-on: ubuntu-latest
+    permissions:
+      id-token: write
+    environment:
+      name: pypi
+      url: https://pypi.org/project/rlm-code/
+    steps:
+      - name: Download distributions
+        uses: actions/download-artifact@v4
+        with:
+          name: release-dist
+          path: dist/
+
+      - name: Publish
+        uses: pypa/gh-action-pypi-publish@release/v1

.gitignore

@@ -85,8 +85,8 @@ target/
 profile_default/
 ipython_config.py
 
-# pyenv
-.python-version
+# pyenv (kept for uv — do NOT ignore .python-version)
+# .python-version
 
 # pipenv
 Pipfile.lock

.python-version

@@ -0,0 +1 @@
+3.11

README.md

@@ -99,16 +99,18 @@ See all available benchmarks:
 
 ### 5. View results
 
+Use the **Research** tab (`Ctrl+5`) for live benchmark and trajectory views.  
+After at least two benchmark runs, export a compare report:
+
 ```
-/leaderboard
+/rlm bench report candidate=latest baseline=previous format=markdown
 ```
 
-Shows a table of all your benchmark runs ranked by reward score.
-
 ### 6. Replay a session step-by-step
 
 ```
-/rlm replay
+/rlm status
+/rlm replay <run_id>
 ```
 
 Walk through the last run one step at a time — see what code the LLM wrote, what output it got, and what it did next.
@@ -134,10 +136,12 @@ This means the LLM can handle documents much larger than its context window, bec
 | `/connect <provider> <model>` | Connect to an LLM |
 | `/model` | Interactive model picker |
 | `/status` | Show connection status |
+| `/sandbox profile secure` | Apply secure sandbox defaults (Docker-first + strict pure RLM) |
 | `/rlm run "<task>"` | Run a task through the RLM loop |
 | `/rlm bench preset=<name>` | Run a benchmark preset |
 | `/rlm bench list` | List available benchmarks |
-| `/leaderboard` | View benchmark results |
+| `/rlm bench compare` | Compare latest benchmark run with previous run |
+| `/harness run "<task>"` | Run tool-using coding harness loop |
 | `/rlm replay` | Step through the last run |
 | `/rlm chat "<question>"` | Ask the LLM a question about your project |
 | `/help` | Show all available commands |
@@ -164,18 +168,26 @@ This means the LLM can handle documents much larger than its context window, bec
 Create an `rlm_config.yaml` in your project directory to customize settings:
 
 ```yaml
-rlm:
-  paradigm: pure_rlm       # pure_rlm, codeact, or traditional
-  max_steps: 30             # max REPL iterations per run
-  timeout: 60               # seconds
+name: my-project
 
-  sandbox:
-    runtime: local          # local, docker, modal, e2b, daytona
+models:
+  openai_api_key: null
+  openai_model: gpt-5.3-codex
 
-  mcp_server:
-    enabled: false
-    transport: stdio
-    port: 8765
+default_model: gpt-5.3-codex
+
+sandbox:
+  runtime: docker
+  superbox_profile: secure
+  superbox_auto_fallback: true
+  superbox_fallback_runtimes: [docker, daytona, e2b]
+  pure_rlm_backend: docker
+  pure_rlm_strict: true
+  pure_rlm_allow_unsafe_exec: false
+
+rlm:
+  default_benchmark_preset: dspy_quick
+  benchmark_pack_paths: []
 ```
 
 Or generate a full sample config:
@@ -202,7 +214,7 @@ rlm_code/
   mcp/              # MCP server for tool integration
   models/           # LLM provider adapters
   sandbox/          # Sandboxed code execution
-  observability/    # MLflow, OpenTelemetry, LangSmith, LangFuse
+  harness/          # Tool-using coding harness (/harness)
 ```
 
 ## Documentation

add_numbers.py

@@ -0,0 +1,11 @@
+def add_numbers(a: float, b: float) -> float:
+    """Add two numbers and return the result."""
+    return a + b
+
+
+if __name__ == "__main__":
+    # Example usage
+    num1 = 5
+    num2 = 3
+    result = add_numbers(num1, num2)
+    print(f"{num1} + {num2} = {result}")