dsh-quality-score
Run the following command in DeepSeek Harness:
dsh plugin install zoahdev/dsh-quality-score
Paste the following prompt into your AI chat to install this plugin:
Install the plugin by running dsh plugin install zoahdev/dsh-quality-score in your DeepSeek Harness terminal; the source repository is at https://github.com/zoahdev/dsh-quality-score .
About this plugin
The dsh registry now hosts over 900 plugins, and ecosystem health scans and recommendation systems keep evolving, yet there was no lightweight tool that could score a plugin from npm metadata alone and point out exactly what to fix. dsh-quality-score fills that gap by distilling the supply-chain health-scan methodology into a single 0-100 scorecard that makes each plugin quality state immediately visible.
The score covers six dimensions: manifest completeness, peer-range resolvability, dist-tag consistency (the #2763 class), dead-range detection, release freshness, and dsh-tools peer compatibility. Each dimension has a fixed maximum contribution, and the total maps to an A-through-F grade accompanied by concrete issues and fix suggestions. A --batch mode turns any list of package names into a ranked leaderboard, making it the natural building block for a registry-wide quality table.
Zero runtime dependencies, read-only by design, and gracefully degradable offline with no API key required. It is ideal for plugin authors auditing their own package, registry maintainers spotting systemic problems, and dsh agents calling the score inline during a conversation to return a structured dsh-quality-score/v1 report in one shot.
Use Cases
- Plugin authors run a quality self-audit before publishing and get actionable fix suggestions
- Registry maintainers use --batch to score all packages and generate a ranked leaderboard
- dsh agents invoke the score inline in a conversation and receive a structured v1 report
Best For
- dsh plugin developers and contributors
- Registry and ecosystem maintenance teams
- End users operating plugins through dsh agents
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.