.forgejo/workflows/ci.yml

@@ -6,24 +6,6 @@ on:
   pull_request:
 
 jobs:
-  lint:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-
-      - uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
-
-      - name: Install package with dev extras
-        run: pip install -e ".[dev]"
-
-      - name: Ruff (lint)
-        run: ruff check src/ tests/
-
-      - name: Mypy (type check)
-        run: mypy
-
   test:
     runs-on: ubuntu-latest
     strategy:

docs/SECURITY_AND_OPERATIONS.md

@@ -34,8 +34,7 @@ testbar von der Orchestrierung.
 
 1. **Eingebaute Defaults** (`config.builtin_defaults()`) — spiegeln die
    Standardwerte der ursprünglichen Shell-Skripte wider (Image
-   `ghcr.io/ggml-org/llama.cpp` per Digest gepinnt für Reproduzierbarkeit,
-   `host_port=8001`,
+   `ghcr.io/ggml-org/llama.cpp:server-cuda`, `host_port=8001`,
    `container_name=va_llm`, `gpu_device=1`, `jinja/fa/kv_unified/
    cont_batching/no_context_shift=true`, `reasoning=on`,
    `cache_type_k/v=q4_0`, `batch_size=1024`, `ubatch_size=512`,

llama.cpp.config.example

@@ -14,9 +14,7 @@
 # locking (--change), and --stop/--check targeting.
 
 [default]
-# Pinned by digest for reproducibility. The :server-cuda tag is a moving target;
-# to update, pull it, read the new digest, and replace the pin below.
-image = ghcr.io/ggml-org/llama.cpp@sha256:5535de118ed457f761cbfeacd7e10fef31cb391ca7cac1d5c78b11d28fcf88e6
+image = ghcr.io/ggml-org/llama.cpp:server-cuda
 # hf_home unterstützt Environment-Variablen und ~, z. B. hf_home = ${HF_HOME}
 hf_home = /srv/models
 model_path = qwen3/default.gguf

pyproject.toml

@@ -17,9 +17,6 @@ dependencies = [
 [project.optional-dependencies]
 dev = [
     "pytest>=7.4",
-    "ruff>=0.6",
-    "mypy>=1.11",
-    "types-requests",
 ]
 
 [project.scripts]
@@ -30,13 +27,3 @@ where = ["src"]
 
 [tool.setuptools.package-data]
 llamacppctl = ["py.typed"]
-
-[tool.ruff]
-target-version = "py310"
-line-length = 100
-
-[tool.mypy]
-python_version = "3.10"
-files = ["src/llamacppctl"]
-warn_unused_ignores = true
-warn_redundant_casts = true

requirements-dev.txt

@@ -2,6 +2,3 @@
 # in pyproject.toml).
 -r requirements.txt
 pytest>=7.4
-ruff>=0.6
-mypy>=1.11
-types-requests

src/llamacppctl/config.py

@@ -36,10 +36,7 @@ class ConfigError(ValueError):
 def builtin_defaults() -> dict:
     """Mirrors the original start-llm-server.sh / status-llm-server.sh defaults."""
     return {
-        # Pinned by digest for reproducibility (the :server-cuda tag is a moving
-        # target). To update: docker pull ghcr.io/ggml-org/llama.cpp:server-cuda,
-        # read the new digest, and bump it here (or override `image` in the config).
-        "image": "ghcr.io/ggml-org/llama.cpp@sha256:5535de118ed457f761cbfeacd7e10fef31cb391ca7cac1d5c78b11d28fcf88e6",
+        "image": "ghcr.io/ggml-org/llama.cpp:server-cuda",
         "hf_home": "/models",
         "model_path": "qwen3/default.gguf",
         "container_name": "va_llm",

src/llamacppctl/docker_ops.py

@@ -11,6 +11,7 @@ import shlex
 import subprocess
 from dataclasses import dataclass
 from pathlib import Path
+from typing import Optional
 
 from .schema import ServerConfig
 

src/llamacppctl/lock_ops.py

@@ -8,7 +8,6 @@ from __future__ import annotations
 
 import fcntl
 from pathlib import Path
-from typing import TextIO
 
 
 class LockError(RuntimeError):
@@ -18,7 +17,7 @@ class LockError(RuntimeError):
 class FileLock:
     def __init__(self, path: Path):
         self.path = Path(path)
-        self.fd: TextIO | None = None
+        self.fd = None
 
     def __enter__(self) -> "FileLock":
         self.path.parent.mkdir(parents=True, exist_ok=True)

src/llamacppctl/prompt_io.py

@@ -202,10 +202,7 @@ def _pin_dns(host: str, allowed_ips: list):
                 )
         return results
 
-    # Intentional monkeypatch: pin DNS to the pre-validated addresses for the
-    # duration of the request (SSRF/rebinding defense). Signature differs from
-    # the stdlib function, hence the targeted ignore.
-    socket.getaddrinfo = pinned  # type: ignore[assignment]
+    socket.getaddrinfo = pinned
     try:
         yield
     finally: