Plain text model image recognition plug-in
whitelonng/dsh-plugin-describe-image · skills
DeepSeek Harness plugin: describeimage — give a text-only model vision through an OpenAI-compatible VLM endpoint.
★ 6GitHub stars
0Forks
2026-08-15Last updated
TypeScriptLanguage
MITLicense
Key features
- Live configuration cardthe Web GUI's Settings → Plugins → "Image understanding" card edits baseURL, model, and the API key (via the credential seam) with immediate effect — no restart.
- LINUX DO — This project is continuously shared and discussed in the LINUX DO community.
- Per-call API key resolutioninline apiKey → credential seam (apiKeyEnv, default VISION_API_KEY) → launch environment.
- Security and boundsredirects refused on every request, maxBytes / maxOutputTokens / timeoutMs bounds, magic-byte media-type gate, bounded error excerpts, secrets never logged.
Install command
dsh plugin --profile web add github:whitelonng/dsh-plugin-describe-image