llmcode-cli 1.0.0__tar.gz

llmcode_cli-1.0.0/.github/workflows/ci.yml

@@ -0,0 +1,26 @@
+name: CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.11", "3.12"]
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+      - run: pip install -e ".[dev,websocket]"
+      - run: ruff check llm_code/ tests/
+      - run: pytest --cov=llm_code --cov-report=xml -q
+      - uses: codecov/codecov-action@v4
+        if: matrix.python-version == '3.12'
+        with:
+          file: coverage.xml

llmcode_cli-1.0.0/.github/workflows/publish.yml

@@ -0,0 +1,21 @@
+name: Publish to PyPI
+
+on:
+  release:
+    types: [published]
+
+permissions:
+  id-token: write
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    environment: pypi
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - run: pip install build
+      - run: python -m build
+      - uses: pypa/gh-action-pypi-publish@release/v1

llmcode_cli-1.0.0/.gitignore

@@ -0,0 +1,56 @@
+# Python
+.venv/
+__pycache__/
+*.egg-info/
+.eggs/
+dist/
+build/
+*.pyc
+*.pyo
+
+# Testing
+.coverage
+htmlcov/
+coverage.xml
+.pytest_cache/
+
+# IDE
+.idea/
+.vscode/
+*.swp
+*.swo
+*~
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Environment & Secrets
+.env
+.env.*
+*.key
+*.pem
+
+# llm-code runtime
+.llm-code/config.local.json
+.llm-code/cron.lock
+.llm-code/result_cache/
+.llm-code/recordings/
+.llm-code/checkpoints/
+.llm-code/swarm/
+
+# Node
+node_modules/
+ink-ui/node_modules/
+ink-ui/dist/
+.superpowers/
+
+# Superpowers plans & specs (development-only docs)
+docs/superpowers/
+
+# Test artifacts (from llm-code self-testing)
+test.txt
+test_output.log
+python_tutorial.py
+九九乘法表.py
+*.log

llmcode_cli-1.0.0/CHANGELOG.md

@@ -0,0 +1,83 @@
+# Changelog
+
+## v0.1.0 (2026-04-03) — Production Cleanup
+
+### Changed
+
+- Default CLI is now Ink UI (React/Node.js); use `--lite` for print-based fallback
+- Updated `pyproject.toml` GitHub URLs from placeholder to `adamhong/llm-code`
+- README updated: Ink UI default, `--lite`/`--serve`/`--connect`/`--ssh` flags documented, ClawHub marketplace, cost tracking, model aliasing
+
+### Fixed
+
+- `[send error:]` debug print in `ink_bridge.py` replaced with `logging.debug`
+- Dead code: removed unused `removed` variable in `algorithms/gemma4_agent.py`
+- Bare `except:` in `algorithms/gemma4_agent.py` replaced with `except Exception:`
+- Unused `subargs` variable in `cli/tui.py` `/session` handler removed
+- Semicolons on same-line imports in `ink_bridge.py` split to two statements
+- Test `test_cli/test_image.py` updated: `detect_image_references` aliased to `extract_dropped_images`
+
+### Removed
+
+- `bubble_sort.py`, `multiplication.py`, `simple_demo.py` — development test artifacts
+- `llm_code/algorithms/` directory — unreferenced Gemma4 agent prototype
+
+### Chores
+
+- Ruff lint: 39 issues fixed (34 auto-fixed, 5 manually resolved)
+- All 1089 tests pass (3 skipped)
+
+## v0.1.0 — Initial Release (2026-04-03)
+
+### Features
+
+**Core Agent (v1)**
+- 6 built-in tools: read_file, write_file, edit_file, bash, glob_search, grep_search
+- Multi-provider support: OpenAI-compatible API + Anthropic
+- Dual-track tool calling: native function calling + XML tag fallback
+- Streaming output with Rich Markdown rendering
+- Layered permission system (5 modes + allow/deny lists)
+- Hook system (pre/post tool use with exit code semantics)
+- Session persistence and multi-session switching
+- Layered config (user → project → local → CLI)
+- Vision fallback for non-vision models
+- Context compaction
+
+**Smart Safety (v2)**
+- Input-aware safety classification (bash ls = read-only, rm = destructive)
+- Safety → permission system integration (dynamic effective_level)
+- Pydantic runtime input validation
+- Tool progress streaming via thread pool + asyncio.Queue
+
+**Ecosystem (v3)**
+- MCP client (stdio + HTTP transport, JSON-RPC 2.0)
+- Plugin marketplace (5 registries: Official, Smithery, npm, GitHub, custom)
+- Claude Code plugin.json compatibility
+- Skills system (auto-inject + slash command trigger)
+- Incremental streaming Markdown rendering
+- Prefix cache optimization
+
+**Agent Capabilities (v4)**
+- Sub-agent tool (asyncio.gather parallel execution)
+- Specialized agent roles (Explore, Plan, Verify)
+- Model routing (static config + per-call override)
+- Git-based undo/checkpoint (auto before writes)
+- 7 git-aware tools with sensitive file detection
+
+**Deep Integration (v5)**
+- LSP integration (3 query tools + auto-detect)
+- Cross-session memory (key-value + auto session summaries)
+- Project indexer (file tree + regex symbol extraction)
+
+**Production Quality (v6)**
+- 4-level context compression (snip → micro → collapse → auto)
+- Streaming tool execution (read-only tools execute during model output)
+- Reactive compact (413 error recovery)
+- Token budget and tool result budget
+- MCP server instructions injection
+- Structured logging
+- Graceful shutdown
+- Config validation
+- GitHub Actions CI
+- Docker support
+- Documentation site

llmcode_cli-1.0.0/CONTRIBUTING.md

@@ -0,0 +1,70 @@
+# Contributing to llm-code
+
+## Development Setup
+
+```bash
+git clone https://github.com/user/llm-code
+cd llm-code
+python -m venv .venv && source .venv/bin/activate
+pip install -e ".[dev]"
+```
+
+## Running Tests
+
+```bash
+pytest                          # all tests
+pytest -x                       # stop on first failure
+pytest --cov=llm_code           # with coverage
+pytest tests/test_tools/        # specific directory
+```
+
+## Code Style
+
+- Formatter/linter: `ruff check --fix llm_code/ tests/`
+- Type hints on all function signatures
+- Frozen dataclasses for immutable data
+- Max file size: 400 lines (split if larger)
+
+## Pull Request Process
+
+1. Create a feature branch
+2. Write tests first (TDD)
+3. Run `pytest` and `ruff check`
+4. Submit PR with clear description
+
+## Architecture
+
+```
+cli → runtime → {tools, api}
+         ↓
+      mcp, marketplace, lsp
+```
+
+- `api/` — Provider abstraction (no other deps)
+- `tools/` — Tool system (does not depend on api)
+- `runtime/` — Combines api + tools
+- `cli/` — Only talks to runtime
+
+## Adding a New Tool
+
+1. Create `llm_code/tools/my_tool.py`
+2. Extend `Tool` ABC
+3. Add Pydantic input model
+4. Implement safety methods (is_read_only, is_destructive)
+5. Register in `cli/app.py`
+6. Add tests in `tests/test_tools/`
+
+## Adding an MCP Server
+
+Add to `~/.llm-code/config.json`:
+
+```json
+{
+  "mcpServers": {
+    "my-server": {
+      "command": "npx",
+      "args": ["-y", "my-mcp-server"]
+    }
+  }
+}
+```

llmcode_cli-1.0.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Adam Hong
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.