dsh-proactive
A DeepSeek Harness proactive autonomous scheduler for multi-model task orchestration, learning, and governed execution.
Install
No clear install command was found in the current repository evidence.
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
- Registers 18 tools for scheduling, memory, autonomy, infrastructure, and management tasks.
- Ingests webhook, file-watch, polling, and manually injected signals through a Sentinel layer.
- Makes execute, defer, dismiss, or ask-user decisions using statistical feedback.
- Builds DAG plans and coordinates parallel multi-model execution with quality reflection and retries.
- Maintains persistent memory with SQLite where supported and a JSON atomic-write fallback.
- Provides hybrid memory retrieval using FTS5, sparse vector similarity, and a memory graph.
- Runs an autonomous heartbeat for observation, goal generation, exploration, strategy evolution, and memory maintenance.
- Exposes governance controls including rate limits, budgets, circuit breaking, confidence gates, and a kill switch.
- Can integrate with a host tool registry when @deepseek-ai/dsh-tools is available; otherwise it falls back to an internal registry.
- Includes optional subsystems for consensus, synchronization, hot reload, encryption, tenancy, benchmarking, dashboards, and a cognitive-energy economy.
Useful For
- Automating recurring observation and response workflows in a DSH host.
- Selecting and coordinating configured LLMs for multi-step DAG tasks.
- Retaining task patterns, model profiles, feedback, and distilled strategies across sessions.
- Applying guarded autonomous exploration and policy adjustment under configured budgets and safety controls.
- Monitoring tool failures and optionally governing host tool execution through the host-fusion layer.
Who It Fits
- DeepSeek Harness users operating Cordis-based hosts.
- Developers building autonomous multi-model workflows.
- Teams needing scheduler-oriented memory, observability, and governance features.
Documented Limitations
- Requires Node.js ^22.18.0 or >=24.11.0 and pnpm according to the README.
- Its LLM requests depend on LLM clients and credentials configured or available through the DSH host, environment, or supported local configuration paths.
- Official host tool-registry integration is only active when the host provides @deepseek-ai/dsh-tools; otherwise it uses an internal registry.
- SQLite persistence falls back to a JSON atomic-write backend when encryption is enabled or the host does not support the SQLite path.
- Several autonomy, host-fusion, and symbiosis behaviors are configuration-dependent; the README states symbiosis options are disabled by default.
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.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- beijingwahw/dsh-proactive
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- fd8f1aa · 2026-09-15
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
- 4GitHub stars
- Forks
- 0
- Open issues
- 0
- Last commit
- 2026-09-13
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Multi-agent team collaboration plugin for DeepSeek Harness with DAG task scheduling, mailbox messaging, and an interactive Web UI monitor.
View plugin →Native SQLite-backed local project taskboard plugin and UI for DeepSeek Harness with Agent claim and review flows.
View plugin →Persistent multi-model Agent teams and bounded DAG workflows with GUI controls for DeepSeek Harness.
View plugin →Second-model AI auto-review for DeepSeek Harness approval requests with fail-closed safety and session audit.
View plugin →