OpenSenseNova/SenseNova-Skills / skills/sn-image-base
sn-image-base
VerifiedBase-layer skill for the SenseNova-Skills project, providing low-level APIs for image generation, recognition (VLM), and text optimization (LLM). This skill does not preprocess inputs; it only calls backend services and returns results. This skill is not user-facing and is intended for upper-layer skills only.
scripts/sn_image_base/vlm/vlm_adapter.py
View on GitHub"""Abstract base class for VLM (Vision Language Model) adapters."""
from __future__ import annotations
from abc import ABC, abstractmethod
class VlmAdapter(ABC):
"""Uniform async interface for a single Vision Language Model backend.
Each concrete adapter wraps one LLM endpoint + model combination and
exposes a single :meth:`vision_completion` coroutine. Synchronous
calling is intentionally **not** supported; callers must run inside an
asyncio event loop.
**Client ownership contract** — when a shared
:class:`httpx.AsyncClient` is supplied at construction time the adapter
*reuses* it and must **not** close it; the caller retains full ownership
of the client's lifecycle. When no external client is provided the
adapter creates and owns an internal client and must close it in
:meth:`aclose`.
"""
@abstractmethod
async def vision_completion(
self,
user_prompt: str,
images: list[str | bytes],
system_prompt: str = "",
model: str | None = None,
) -> str:
"""Send image(s) and a text prompt to the model; return the reply.
Args:
user_prompt: User-facing text instruction.
images: One or more images to pass to the model. Each element
is either a file-path string or raw image bytes.
system_prompt: System-level instruction prepended to the
conversation. Defaults to ''.
model: Model name to use. If None, uses the default set at
initialization.
Returns:
str: Raw text response from the model (may contain JSON or
markdown-wrapped JSON depending on the model and prompt).
"""
@abstractmethod
async def aclose(self) -> None:
"""Release async resources owned by this adapter.
Must be called when the adapter is no longer needed. Adapters that
were given an external shared client must implement this as a no-op;
adapters that created their own internal client must close it here.
"""
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |