PC Configuration Plan Generator

Problem it solves

PC build recommendations often get stuck on a mix of budget, brand preference, and use case. Relying on experience alone can miss compatibility risks, while a single configuration tier makes it hard to compare upgrade costs. If the result is only text, it is less useful for screenshots, reviews, and side-by-side discussion.

PC Configuration Plan Generator is aimed at hardware purchase scenarios. Given a budget range, brand preference, and primary use case, such as video editing, gaming, office work, or 3D rendering, it generates high, medium, and low tiers of builds and outputs a fully Chinese, tech-style comparison image for direct sharing and comparison.

Core capabilities and workflow

The core workflow connects requirement parsing, market data lookup, configuration inference, and compatibility checks:

• Deep intent parsing: extracts budget range, brand preference, main use case, and implied hardware requirements from natural language.
• Hardware market lookup: generates search keywords and uses current hardware market information to infer configurations.
• Three-tier inference: outputs high, medium, and low configuration lists rather than a single recommendation.
• Compatibility validation: checks component conflicts, outputs compatibility ratings, corrected components, and upgrade paths.
• Visual generation: combines configuration details, compatibility notes, and purchase suggestions into an image prompt, then generates a comparison image at the requested aspect ratio.

In real execution, the skill uses asynchronous tasks: submit a job to get a task_id, then poll for the result. Without an appKey, it enters a simulated call for quick preview; this is not real generated content.

Boundaries and notes

It is useful for initial screening, side-by-side comparison, and presentation of PC configurations, especially when turning budget, use case, and brand preference into a readable comparison image.

Things to note:

• requirement_desc is optional, but specific budget, use case, and brand preference improve inference stability.
• Real tasks are strictly asynchronous; synchronous waiting is not supported.
• During polling, each round's core data should be shared with the user, not just the final result.
• Compatibility ratings and purchase suggestions are model-derived; final purchases should still follow physical specs, official documentation, and warranty terms.
• Simulated results are for preview only; real usage requires configuring a ShowAPI AppKey.