DeepSeek Harness PluginManifest Valid

dsh-autoresearch

Durable auto-research and iterative experiment loop plugin for DeepSeek Harness with Web GUI monitoring.

Developer Tools

0GitHub Stars0ForksUpdated2026-09-29

Source

Install

$ dsh plugin --profile web add github:aa2246740/dsh-autoresearch#v1.0.10

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-autoresearch is a workflow plugin for DeepSeek Harness that introduces a durable, iterative research and experiment loop. It empowers agents to modify project files, measure metric outcomes, retain improvements, or roll back failed iterations, with full experiment tracking exposed in the official Web and Desktop GUI.

Key Capabilities

Autonomous research loops that iteratively modify code, measure metrics, and decide whether to keep or revert changes
Local baseline snapshot protection with automated rollback on failures, crashes, or discarded steps
File-backed experiment ledger stored in the project's .auto/ directory tracking prompts, scripts, logs, and ideas
Interactive slash commands including /autoresearch, /autoresearch resume, status, off, and clear
Integration into the official DeepSeek Harness Web GUI with dedicated status and monitoring panels

Useful For

Automating performance tuning and benchmark-driven code optimization
Running multi-round empirical experiments and hypothesis validation in a project
Iterative exploration of algorithmic improvements with automated rollback safety

Who It Fits

Developers conducting iterative optimization and refactoring tasks in DeepSeek Harness
Researchers exploring automated parameter or prompt experiments within a local workspace

Documented Limitations

Requires DeepSeek Harness host version matching >=0.2.0-rc.1 <0.2.1 and Node.js >=22.19.0
Directly modifies project files and executes local test/measurement shell scripts
Multi-round continuous execution consumes LLM model token quota

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, 21 development, 14 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: aa2246740/dsh-autoresearch
Registry source: GitHub · dsh-plugin topic
Registry classification: Plugin
Source checked: d58a1fd · 2026-10-22

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.