← DeepSeek Harness Plugins
DeepSeek Harness PluginIndexed

better-dsh

Enhances DeepSeek Harness with persistent IPython REPL, URL schemas, agent swarms, and context revive.

UI & ProductivityDeveloper ToolsBrowser & WebSkills & Workflows
6GitHub Stars0ForksUpdated2026-10-06

Install

$ dsh plugin --profile web add --config.auto-install-peers=false better-dsh@latest

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
better-dsh (Dashr) is an enhancement plugin for DeepSeek Harness (dsh) designed for serious coding workflows. It introduces a persistent IPython REPL kernel, unified runtime URL schemas (skill://, ctx://, agent://, dvc://, dsh://), context-as-variables support, dill-based kernel state revival across restarts, KeepRecency context compaction, LLM endpoint failover, and browser device integrations.

Key Capabilities

  • Persistent IPython REPL kernel replacing ephemeral execution environments
  • Unified resource URL schemas including skill://, ctx://, agent://, dvc://, and dsh://
  • Context as Variables allowing transcripts and sessions to be accessed programmatically
  • Full context revive using dill snapshots across restarts
  • Subagent, workflow, and Ralph loop first-class invocations
  • KeepRecency context compactions targeting 50% usable context threshold
  • Automatic global LLM endpoint failover for unattended operations
  • Browser device control via dvc://browser and iOS PWA UI enhancements

Useful For

  • Executing complex multi-step coding workflows with persistent Python state
  • Composing subagents, browser automation, and local context into unified REPL cells
  • Maintaining session continuity and namespace persistence across DSH restarts
  • Managing long-running autonomous tasks with LLM endpoint failover protection

Who It Fits

  • Software engineers using DeepSeek Harness for advanced coding tasks
  • AI agent developers creating multi-agent and workflow automation loops
  • Power users requiring persistent Python kernels and context inspection

Documented Limitations

  • Requires an environment capable of running IPython and Python dill snapshots
  • Changes context compaction defaults to 50% threshold which may trigger earlier compactions

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.

License Declared

GitHub reports the repository license as MIT.

info
Package Manifest Not Found

No root package.json was captured in the latest GitHub snapshot.

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
pgmi-builds/better-dsh
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
fe3c99a · 2026-10-09

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
6GitHub stars
Forks
0
Open issues
0
Last commit
2026-10-06
Last release
No release detected
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