subochev/routerai-image-mcp
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

routerai-image-mcp: MCP-сервер генерации картинок через RouterAI

Browse Source 
- server.py: MCP-сервер (stdio + streamable HTTP), инструменты generate_image / list_image_models
- два бэкенда: chat/completions (Gemini Image, понимает правку по референсу) и images/generations (FLUX, GPT-Image)
- ключ RouterAI берётся из env или ~/.hermes/models.yaml автоматически
- retry на 429/5xx, картинка сохраняется файлом (~/.hermes/image_cache/mcp-gen)
- test_client.py / test_edit.py: проверка по протоколу, включая image-to-image
- README: установка, настройка, подключение к Hermes/opencode, питфоллы
This commit is contained in:
Субочев Антон
2026-09-24 11:01:23 +03:00
commit 585c6a4ca4
7 changed files with 793 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.gitignore
+8
View File
@@ -0,0 +1,8 @@
__pycache__/
*.pyc
.venv/
*.png
*.jpg
*.jpeg
*.webp
.env
LICENSE
+202
View File
@@ -0,0 +1,202 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
README.md
+222
View File
@@ -0,0 +1,222 @@
# routerai-image-mcp
MCP-сервер генерации и редактирования картинок через **RouterAI**
(`https://routerai.ru/api/v1`). Даёт любой модели — в том числе той, что
рисовать не умеет, но умеет ставить ТЗ другим — инструмент `generate_image`:
передал текстовое описание, получил файл с картинкой.
## Зачем
Агентные модели (DeepSeek, Qwen, локальные) отлично формулируют ТЗ, но не
генерируют изображения. Этот сервер закрывает разрыв: модель вызывает
`generate_image(prompt=...)`, сервер ходит в модель генерации картинок
(Nano Banana / Gemini Image, FLUX, GPT-Image), кладёт результат файлом и
возвращает путь. Модель вставляет этот путь в ответ — клиент показывает картинку.
## Архитектура
```
LLM (agent) ──MCP──▶ routerai-image-mcp ──HTTPS──▶ routerai.ru/api/v1
│
└─▶ файл на диске (~/.hermes/image_cache/mcp-gen/)
возвращает: "MEDIA:<путь>"
```
Два бэкенда, выбираются автоматически по id модели:
| Backend | Что вызывается | Модели |
|---|---|---|
| `chat` | `POST /chat/completions` с `modalities: ["image","text"]` | Gemini Image / Nano Banana |
| `images` | `POST /images/generations` | FLUX.2, GPT-Image |
`chat`-бэкенд дополнительно умеет **image-to-image**: если передать
`reference_images`, они уходят в запрос как image_url-части, и модель
редактирует присланную картинку (сохранив сюжет) — проверено на «поменяй фон».
## Требования
- Python 3.10+
- Пакет `mcp` (`pip install mcp`) — в Hermes он уже есть:
`/usr/local/lib/hermes-agent/venv/bin/python`
- Ключ RouterAI (см. «Настройка»)
- Сетевой доступ к `https://routerai.ru`
## Установка
```bash
git clone https://git.binom.pw/subochev/routerai-image-mcp
cd routerai-image-mcp
python3 -m venv .venv
.venv/bin/pip install -r requirements.txt
```
Либо без своего venv — использовать интерпретатор Hermes (там `mcp` уже стоит):
```bash
/usr/local/lib/hermes-agent/venv/bin/python server.py
```
## Настройка
Ключ ищется в таком порядке:
1. `ROUTERAI_API_KEY` (или `IMAGE_MCP_API_KEY`) в окружении;
2. `~/.hermes/models.yaml` → `models.deepseek.api_key` (если у записи
`base_url` содержит `routerai.ru`);
3. любой подходящий `api_key` из `models.yaml`, где `base_url` = routerai.ru.
То есть на машине с Hermes, где `models.yaml` уже настроен, **ничего задавать
не нужно** — ключ подхватится сам. Явный способ:
```bash
export ROUTERAI_API_KEY='sk-...'
```
Переменные окружения сервера:
| Переменная | Дефолт | Смысл |
|---|---|---|
| `ROUTERAI_API_KEY` | — | Ключ RouterAI (иначе берётся из `models.yaml`) |
| `ROUTERAI_BASE_URL` | `https://routerai.ru/api/v1` | Базовый URL API |
| `IMAGE_MCP_DEFAULT_MODEL` | `google/gemini-3.1-flash-image` | Модель по умолчанию |
| `IMAGE_MCP_OUT_DIR` | `~/.hermes/image_cache/mcp-gen` | Куда складывать картинки |
| `IMAGE_MCP_TIMEOUT` | `300` | Таймаут запроса, сек |
| `MCP_TRANSPORT` | `stdio` | `stdio` или `http` |
| `MCP_HOST` / `MCP_PORT` | `0.0.0.0` / `8790` | Только для `http` |
## Запуск
### stdio (обычный режим для локальных клиентов)
```bash
python server.py
```
### HTTP (когда сервер нужен по сети нескольким клиентам)
```bash
MCP_TRANSPORT=http MCP_PORT=8790 python server.py
```
Эндпоинт: `http://<host>:8790/mcp` (streamable HTTP). Проверка:
```bash
curl -s -X POST http://127.0.0.1:8790/mcp \
-H 'Content-Type: application/json' \
-H 'Accept: application/json, text/event-stream' \
-d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"probe","version":"1"}}}'
# → event: message / data: {... "serverInfo":{"name":"routerai-image"}}
```
## Подключение к Hermes
```bash
hermes mcp add routerai-image \
--command /usr/local/lib/hermes-agent/venv/bin/python \
--args /opt/routerai-image-mcp/server.py
```
На вопрос «Enable all N tools?» ответить `Y`. Проверка:
```bash
hermes mcp test routerai-image
# ✓ Connected, Tools discovered: 2 → generate_image, list_image_models
```
⚠️ **Инструменты появляются только в новой сессии** — MCP-серверы
подключаются при старте Hermes; в уже идущем диалоге их не будет.
Если сервер развёрнут на отдельной машине, вместо stdio удобнее HTTP:
```bash
hermes mcp add routerai-image --url http://<host>:8790/mcp
```
## Подключение к другим клиентам
opencode (`~/.config/opencode/config.json`):
```json
{
"mcp": {
"routerai-image": {
"type": "local",
"command": ["/usr/local/lib/hermes-agent/venv/bin/python", "/opt/routerai-image-mcp/server.py"],
"enabled": true
}
}
}
```
Claude Desktop / прочие stdio-клиенты — тот же `command` + `args`.
## Инструменты
### `generate_image`
| Аргумент | Тип | Смысл |
|---|---|---|
| `prompt` | str | ТЗ: сюжет, стиль, свет, композиция, пропорции |
| `model` | str | id модели RouterAI (дефолт `google/gemini-3.1-flash-image`) |
| `reference_images` | list[str] | Пути к картинкам для правки/референса (image-to-image) |
| `out_dir` | str | Куда сложить файл |
| `filename` | str | Имя файла (расширение подставится само) |
Возвращает текст с абсолютным путём и строкой `MEDIA:<путь>`. **Чтобы клиент
показал картинку, модель должна вставить эту строку `MEDIA:<путь>` целиком в
свой ответ.**
### `list_image_models`
Печатает список поддерживаемых моделей, их бэкенды и дефолт.
## Модели и стоимость
| id | Что это | Цена (факт, за картинку) |
|---|---|---|
| `google/gemini-3.1-flash-image` | Nano Banana 2 — дефолт | ~7.5 ₽ |
| `google/gemini-3.1-flash-lite-image` | Nano Banana 2 Lite — самый дешёвый | ~3.7 ₽ |
| `google/gemini-2.5-flash-image` | Nano Banana (2.5) | ~4.3 ₽ |
| `google/gemini-3-pro-image` | Nano Banana Pro | ~13 ₽ |
| `black-forest-labs/flux.2-klein-4b` | FLUX.2 Klein — дешёвый, без правки | — |
| `black-forest-labs/flux.2-pro` / `-max` / `-flex` | FLUX.2 Pro/Max/Flex | — |
| `openai/gpt-image-1` / `-mini` | GPT Image | — |
Ключи без значения — считаются по своему тарифу, проверяйте на
`GET https://routerai.ru/api/v1/models` (поле `pricing.image_output`).
Gemini-Image дороже других, зато понимает сложные ТЗ и умеет правку по референсу.
## Проверка
```bash
# список инструментов по протоколу + генерация
/usr/local/lib/hermes-agent/venv/bin/python test_client.py --generate google/gemini-3.1-flash-lite-image
# правка существующей картинки по референсу
/usr/local/lib/hermes-agent/venv/bin/python test_edit.py /путь/к/картинке.jpg
```
Оба скрипта поднимают сервер по stdio и работают с ним как настоящий MCP-клиент.
## Питфоллы
- **`llm.binom.pw` может лежать (502), а `routerai.ru` — работать.** Сервер
ходит напрямую в `routerai.ru` и не зависит от Bifrost/llm-proxy. Если у вас
всё настроено через роутер и он упал — картинки всё равно будут.
- **Путь к интерпретатору.** `python3` из системы может не иметь пакета `mcp`.
В Hermes надёжнее указывать его venv: `/usr/local/lib/hermes-agent/venv/bin/python`.
- **Картинки нужны модели, а не агенту.** Возвращается путь к файлу, а не
base64 в контекст — контекст не забивается мегабайтами.
- **Модели, отдающие картинки иначе.** `flux.2-*` и `gpt-image-*` не принимают
`modalities` — для них используется `/images/generations`; определяется
автоматически по таблице `MODELS` в начале `server.py`. Новая модель из
другого семейства — дописать её туда.
- **RouterAI отдаёт префикс-мусор перед JSON** на `/images/generations`
(переводы строк и пробелы) — парсер ищет первый `{`.
- **429/5xx** — сервер сам делает 3 повтора с нарастающей паузой; 403
`model_blocked` означает, что модель не разрешена ключу на роутере.
## Лицензия
Apache License 2.0 — см. `LICENSE`.
requirements.txt
+1
View File
@@ -0,0 +1 @@
mcp>=1.2.0
server.py
+289
View File
@@ -0,0 +1,289 @@
#!/usr/bin/env python3
"""RouterAI Image MCP server.
Даёт ЛЮБОЙ модели (в т.ч. той, что не умеет рисовать) инструмент
`generate_image` — генерация/редактирование картинок через RouterAI
(Nano Banana / Gemini Image, FLUX, GPT-Image).
Транспорт:
MCP_TRANSPORT=stdio (по умолчанию) | http
MCP_PORT=8790 (для http)
Ключ ищем в порядке:
1) env ROUTERAI_API_KEY / IMAGE_MCP_API_KEY
2) ~/.hermes/models.yaml -> models.deepseek.api_key (base_url = routerai.ru)
3) ~/.hermes/models.yaml -> любой api_key с base_url routerai.ru
"""
from __future__ import annotations
import base64
import json
import os
import re
import sys
import time
import mimetypes
import urllib.error
import urllib.request
from pathlib import Path
from mcp.server.fastmcp import FastMCP
BASE_URL = os.environ.get("ROUTERAI_BASE_URL", "https://routerai.ru/api/v1").rstrip("/")
DEFAULT_MODEL = os.environ.get("IMAGE_MCP_DEFAULT_MODEL", "google/gemini-3.1-flash-image")
DEFAULT_OUT_DIR = os.environ.get(
"IMAGE_MCP_OUT_DIR", str(Path.home() / ".hermes" / "image_cache" / "mcp-gen")
)
REQUEST_TIMEOUT = int(os.environ.get("IMAGE_MCP_TIMEOUT", "300"))
# модель -> (бэкенд, человеческое имя). backend "chat" = OpenAI chat с modalities,
# backend "images" = POST /images/generations
MODELS: dict[str, tuple[str, str]] = {
"google/gemini-3.1-flash-image": ("chat", "Nano Banana 2 (Gemini 3.1 Flash Image)"),
"google/gemini-3.1-flash-lite-image": ("chat", "Nano Banana 2 Lite"),
"google/gemini-2.5-flash-image": ("chat", "Nano Banana (Gemini 2.5 Flash Image)"),
"google/gemini-3-pro-image": ("chat", "Nano Banana Pro (Gemini 3 Pro Image)"),
"google/gemini-3.1-flash-image-preview": ("chat", "Nano Banana 2 (preview)"),
"google/gemini-3-pro-image-preview": ("chat", "Nano Banana Pro (preview)"),
"openai/gpt-image-1": ("images", "GPT Image 1"),
"openai/gpt-image-1-mini": ("images", "GPT Image 1 Mini"),
"black-forest-labs/flux.2-pro": ("images", "FLUX.2 Pro"),
"black-forest-labs/flux.2-max": ("images", "FLUX.2 Max"),
"black-forest-labs/flux.2-flex": ("images", "FLUX.2 Flex"),
"black-forest-labs/flux.2-klein-4b": ("images", "FLUX.2 Klein 4B"),
}
mcp = FastMCP("routerai-image")
# --------------------------------------------------------------------------- key
def _api_key() -> str:
for env in ("ROUTERAI_API_KEY", "IMAGE_MCP_API_KEY"):
v = os.environ.get(env)
if v and len(v) > 20:
return v.strip()
models_yaml = Path.home() / ".hermes" / "models.yaml"
if models_yaml.exists():
try:
import yaml # type: ignore
data = yaml.safe_load(models_yaml.read_text(encoding="utf-8")) or {}
cands = []
for name, entry in (data.get("models") or {}).items():
if not isinstance(entry, dict):
continue
key = str(entry.get("api_key") or "")
base = str(entry.get("base_url") or "")
if "routerai.ru" in base and key.startswith("sk-"):
cands.append((name != "deepseek", key))
if cands:
cands.sort()
return cands[0][1]
except Exception as exc: # noqa: BLE001
print(f"[routerai-image] models.yaml parse failed: {exc}", file=sys.stderr)
raise RuntimeError(
"RouterAI API key not found. Set ROUTERAI_API_KEY, or put a routerai.ru "
"entry with api_key into ~/.hermes/models.yaml"
)
# ---------------------------------------------------------------------------- http
def _post(path: str, body: dict, timeout: int = REQUEST_TIMEOUT) -> dict:
key = _api_key()
data = json.dumps(body, ensure_ascii=False).encode("utf-8")
last_err = ""
for attempt in range(4):
req = urllib.request.Request(
f"{BASE_URL}{path}",
data=data,
headers={
"Content-Type": "application/json",
"Authorization": f"Bearer {key}",
"User-Agent": "routerai-image-mcp/1.0",
},
)
try:
with urllib.request.urlopen(req, timeout=timeout) as resp:
raw = resp.read()
# RouterAI иногда отдаёт мусорные префиксы перед JSON
start = raw.find(b"{")
return json.loads(raw[start:] if start > 0 else raw)
except urllib.error.HTTPError as exc:
payload = exc.read().decode("utf-8", "replace")[:400]
last_err = f"HTTP {exc.code}: {payload}"
if exc.code in (429, 500, 502, 503, 504) and attempt < 3:
time.sleep(3 * (attempt + 1))
continue
raise RuntimeError(last_err) from None
except Exception as exc: # noqa: BLE001
last_err = f"{type(exc).__name__}: {exc}"
if attempt < 3:
time.sleep(3 * (attempt + 1))
continue
raise RuntimeError(last_err) from None
raise RuntimeError(last_err)
# ------------------------------------------------------------------------- helpers
def _data_url(path: str) -> str:
p = Path(path).expanduser()
if not p.exists():
raise RuntimeError(f"reference image not found: {p}")
mime = mimetypes.guess_type(p.name)[0] or "image/png"
return f"data:{mime};base64,{base64.b64encode(p.read_bytes()).decode()}"
def _decode_image(item: dict) -> bytes:
"""Принимает {'b64_json': ...} или {'url': 'data:...'} / http url."""
if item.get("b64_json"):
return base64.b64decode(item["b64_json"])
url = item.get("url") or (item.get("image_url") or {}).get("url") or ""
if url.startswith("data:"):
return base64.b64decode(url.split(",", 1)[1])
if url.startswith("http"):
with urllib.request.urlopen(url, timeout=REQUEST_TIMEOUT) as resp: # noqa: S310
return resp.read()
raise RuntimeError("no image payload in response")
def _ext_from_bytes(blob: bytes) -> str:
if blob[:8] == b"\x89PNG\r\n\x1a\n":
return ".png"
if blob[:3] == b"\xff\xd8\xff":
return ".jpg"
if blob[:4] == b"RIFF" and blob[8:12] == b"WEBP":
return ".webp"
return ".png"
def _slug(text: str, limit: int = 40) -> str:
txt = re.sub(r"[^\w\s-]", "", text, flags=re.UNICODE).strip()
txt = re.sub(r"[\s-]+", "-", txt)
return (txt[:limit] or "image").strip("-").lower()
def _run(model: str, prompt: str, refs: list[str] | None, out_dir: str, filename: str | None) -> str:
if model not in MODELS:
known = ", ".join(sorted(MODELS))
raise RuntimeError(f"unknown model '{model}'. Known: {known}")
backend, human = MODELS[model]
refs = refs or []
started = time.time()
if backend == "chat":
if refs:
content: list[dict] = [{"type": "text", "text": prompt}]
content += [
{"type": "image_url", "image_url": {"url": _data_url(p)}} for p in refs
]
else:
content = prompt # type: ignore[assignment]
body = {
"model": model,
"messages": [{"role": "user", "content": content}],
"modalities": ["image", "text"],
}
if not refs:
body["response_modalities"] = ["IMAGE"]
data = _post("/chat/completions", body)
msg = (data.get("choices") or [{}])[0].get("message", {}) or {}
images = msg.get("images") or []
if not images and isinstance(msg.get("content"), list):
images = [c for c in msg["content"] if c.get("type") in ("image_url", "image")]
if not images:
raise RuntimeError(f"model returned no image. raw: {json.dumps(data)[:400]}")
blob = _decode_image(images[0])
usage = data.get("usage") or {}
else:
body = {"model": model, "prompt": prompt, "n": 1}
if refs:
body["image"] = _data_url(refs[0])
data = _post("/images/generations", body)
items = data.get("data") or []
if not items:
raise RuntimeError(f"model returned no image. raw: {json.dumps(data)[:400]}")
blob = _decode_image(items[0])
usage = data.get("usage") or {}
out = Path(out_dir).expanduser()
out.mkdir(parents=True, exist_ok=True)
name = filename or f"{time.strftime('%Y%m%d-%H%M%S')}-{_slug(prompt)}"
if not Path(name).suffix:
name += _ext_from_bytes(blob)
dest = out / name
dest.write_bytes(blob)
cost = usage.get("cost")
secs = time.time() - started
cost_line = f"{cost:.2f}₽" if isinstance(cost, (int, float)) else "n/a"
kb = len(blob) / 1024
return (
f"Готово: {human}\n"
f"Файл: {dest} ({kb:.0f} KB)\n"
f"Модель: {model} | время: {secs:.1f}s | стоимость: {cost_line}\n"
f"MEDIA:{dest}\n"
f"(покажи файл пользователю: вставь строку MEDIA:<путь> целиком в ответ)"
)
# --------------------------------------------------------------------------- tools
@mcp.tool()
def generate_image(
prompt: str,
model: str = DEFAULT_MODEL,
reference_images: list[str] | None = None,
out_dir: str = DEFAULT_OUT_DIR,
filename: str | None = None,
) -> str:
"""Сгенерировать картинку по текстовому ТЗ (или отредактировать существующую).
Args:
prompt: ТЗ на картинке — что нарисовать. Чем конкретнее (сюжет, стиль,
свет, композиция, пропорции), тем лучше результат.
model: id модели RouterAI. Дефолт google/gemini-3.1-flash-image
(Nano Banana 2). Дешёвые: google/gemini-3.1-flash-lite-image,
google/gemini-2.5-flash-image. Дорогие: google/gemini-3-pro-image.
Альтернативы: black-forest-labs/flux.2-pro, openai/gpt-image-1.
reference_images: список путей к картинкам-референсам/для правки.
out_dir: куда сложить результат (по умолчанию ~/.hermes/image_cache/mcp-gen).
filename: имя файла без пути (расширение подставится само).
Returns:
Текст с абсолютным путём к файлу и строкой `MEDIA:<путь>` — вставь эту
строку в свой ответ, чтобы клиент показал картинку.
"""
return _run(model, prompt, reference_images, out_dir, filename)
@mcp.tool()
def list_image_models() -> str:
"""Список доступных моделей генерации картинок и их бэкендов."""
lines = [f"{mid} — {human} (backend: {backend})" for mid, (backend, human) in sorted(MODELS.items())]
return "Доступные модели:\n" + "\n".join(lines) + f"\n\nДефолт: {DEFAULT_MODEL}\nБазовый URL: {BASE_URL}"
def main() -> None:
transport = os.environ.get("MCP_TRANSPORT", "stdio").lower()
if transport in ("http", "streamable-http", "sse"):
mcp.settings.host = os.environ.get("MCP_HOST", "0.0.0.0")
mcp.settings.port = int(os.environ.get("MCP_PORT", "8790"))
mcp.run(transport="streamable-http")
else:
mcp.run(transport="stdio")
if __name__ == "__main__":
main()
test_client.py
+41
View File
@@ -0,0 +1,41 @@
#!/usr/bin/env python3
"""Проверка routerai-image-mcp по протоколу stdio: tools/list + tools/call."""
import asyncio, os, sys
from mcp import ClientSession, StdioServerParameters
from mcp.client.stdio import stdio_client
PY = "/usr/local/lib/hermes-agent/venv/bin/python"
SERVER = "/opt/routerai-image-mcp/server.py"
async def main() -> None:
params = StdioServerParameters(
command=PY,
args=[SERVER],
env={**os.environ, "IMAGE_MCP_OUT_DIR": "/tmp/mcp-img-test"},
)
async with stdio_client(params) as (r, w):
async with ClientSession(r, w) as s:
await s.initialize()
tools = await s.list_tools()
print("TOOLS:", [t.name for t in tools.tools])
for t in tools.tools:
print(" -", t.name, "::", (t.description or "").splitlines()[0][:90])
res = await s.call_tool("list_image_models", {})
print("\n--- list_image_models ---")
print(res.content[0].text[:300])
if len(sys.argv) > 1 and sys.argv[1] == "--generate":
model = sys.argv[2] if len(sys.argv) > 2 else "google/gemini-3.1-flash-lite-image"
print(f"\n--- generate_image ({model}) ---")
res = await s.call_tool(
"generate_image",
{"prompt": "рыжий кот в будёновке, акварель, тёплый фон", "model": model},
)
print(res.content[0].text)
print("isError:", res.isError)
asyncio.run(main())
test_edit.py
+30
View File
@@ -0,0 +1,30 @@
#!/usr/bin/env python3
"""Проверка image-to-image: правка существующей картинки по референсу."""
import asyncio, os, sys
from mcp import ClientSession, StdioServerParameters
from mcp.client.stdio import stdio_client
PY = "/usr/local/lib/hermes-agent/venv/bin/python"
SERVER = "/opt/routerai-image-mcp/server.py"
REF = sys.argv[1]
async def main() -> None:
params = StdioServerParameters(
command=PY, args=[SERVER],
env={**os.environ, "IMAGE_MCP_OUT_DIR": "/tmp/mcp-img-test"},
)
async with stdio_client(params) as (r, w):
async with ClientSession(r, w) as s:
await s.initialize()
res = await s.call_tool("generate_image", {
"prompt": "Оставь кота как есть, но поменяй фон на зимний: падает снег, ёлки.",
"model": "google/gemini-3.1-flash-image",
"reference_images": [REF],
"filename": "kot-edit.png",
})
print(res.content[0].text)
print("isError:", res.isError)
asyncio.run(main())