← DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

dsh-prompt-predict

Codex-style composer prompt prediction plugin for DeepSeek Harness that renders ghost text suggestions after assistant replies.

UI & ProductivityDeveloper ToolsBrowser & WebSecurity & Policy
2GitHub Stars0ForksUpdated2026-10-26

Install

$ dsh plugin --profile desktop add [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.

source-grounded
dsh-prompt-predict is an installable DeepSeek Harness host and client plugin. Following each completed assistant response, it predicts the next likely user message and renders it as an inline ghost-text overlay in an empty composer, allowing users to accept the full suggestion with Tab, accept piece by piece with Alt+Right Arrow, request a reroll, or dismiss it with Escape.

Key Capabilities

  • Predicts the user's next message after assistant turns and renders it as composer ghost text
  • Supports multiple interactions: Tab to insert full suggestion, Alt+Right Arrow to accept word-by-word/piece-by-piece, reroll button for a fresh suggestion, and Esc to dismiss
  • Lightweight trimmed context sending only recent user/assistant plain text without tool outputs, attachments, or system prompts
  • Configurable model routing to inherit the session model or pin a custom provider/model
  • Includes settings configuration UI with language selection, context budget tuning, and local prediction history inspection
  • Automatic deduplication and fallback handling when reasoning levels are unsupported

Useful For

  • Accelerating conversational flows and prompt typing during iterative LLM queries
  • Drafting standard follow-up prompts and clarification requests without manual typing
  • Reviewing and partially accepting predicted follow-up directions piece by piece

Who It Fits

  • DeepSeek Harness desktop and web users seeking faster prompt drafting
  • Power users who prefer inline ghost text and keyboard-driven completion workflows

Documented Limitations

  • Requires an empty composer and an idle conversation session to trigger predictions
  • Predictions consume API quota on the designated model
  • Explicitly verified on DeepSeek Harness 0.2.1-rc.3; compatibility with other versions is untested
  • Relies on specific conversation.composer.dock and settings.section UI slot mount points
  • Windows environment is untested

DSH Compatibility

Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.

Not tested yetNo runtime compatibility tests have been published yet.

Security Signals

Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.

Dependency Counts

package.json declares 0 runtime, 0 development, 0 peer, and 0 optional dependencies.

info
Dsh Bundle Declared

package.json declares DSH bundle metadata.

info
License Declared

GitHub reports the repository license as MIT.

info
Package Manifest Available

A root package.json was captured and can be inspected by the registry.

info
Source Available

Public GitHub source metadata is available for this registry snapshot.

info

Source & Registry Notes

Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.

Source repository
ferrisching/dsh-prompt-predict
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
80c4101 · 2026-10-26

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
2GitHub stars
Forks
0
Open issues
0
Last commit
2026-10-26
Last release
2026-10-24
For maintainers

Maintaining this plugin?

This listing is generated from public repository data. If you maintain this project, you can review the information and share this listing with your users if you find it useful.

Add to README
Listed on DSHPlugin.app