dsh-models-input-modalities
Per-model input-modality (text/image) and reasoning level selector for third-party providers in DeepSeek Harness.
Install
$ dsh plugin --profile web add @jcy2387/[email protected]Plugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Adds a per-model input modality selector (Provider default, Text only, Text and image) to third-party provider cards
- Configures custom reasoning effort levels (off, minimal, low, medium, high, xhigh, max) with custom wire spellings per model
- Provides revision-fenced atomic writes back to cordis.patch.yml to prevent concurrent settings conflicts
- Listens to document updates live to synchronize dynamically when models are added or removed
- Imports legacy llm-pi-ai profile settings automatically during migration
- Offers bilingual UI localization in English and Simplified Chinese
Useful For
- Enabling multimodal image input for custom vision-capable third-party LLM endpoints
- Exposing reasoning effort options (low, medium, high, max) in the chat composer model picker for custom reasoning models
- Mapping non-standard gateway reasoning parameter names to supported adapter levels
Who It Fits
- DeepSeek Harness Web users configuring custom third-party OpenAI/Anthropic/compatible AI gateways
- Developers deploying multimodal or reasoning models through external API providers
Documented Limitations
- Dormant/unconfigured provider cards do not render the fold until saved once
- Route-level default settings and built-in model overrides are out of scope and must be configured via YAML
- Does not validate whether the external gateway actually supports the declared reasoning level or input modality
- Cannot save changes in read-only settings deployments
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 1 runtime, 27 development, 16 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as MIT.
A root package.json was captured and can be inspected by the registry.
package.json declares a prepare lifecycle script that may run during relevant package installation workflows.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- DamonBao/dsh-models-input-modalities
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- d32d6bb · 2026-09-30
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 1GitHub stars
- Forks
- 0
- Open issues
- 0
- Last commit
- 2026-09-29
- Last release
- 2026-09-24
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Agent skill for DeepSeek Harness that turns ideas, plans, and codebases into interactive architecture diagrams.
View plugin →Persistent multi-model agent teams with dynamic DAG orchestration, run monitoring, and token insights for DeepSeek Harness.
View plugin →Full office suite integration for DeepSeek Harness supporting spreadsheets, docs, slides, bases, and canvases.
View plugin →Session lifecycle manager for DeepSeek Harness with search, restore, idle cleanup, and diagnosis.
View plugin →