dsh-Visual-Workflow
Visual Agent orchestration and meta-orchestration platform for DeepSeek Harness with human-in-the-loop and organizational memory.
Install
$ dsh plugin --profile web add "github:GZX2211/dsh-Visual-Workflow#main"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
- Visual workflow canvas supporting role, file, database, stage, and collaboration group (Agent Team) nodes
- Meta-orchestration allowing agents to dynamically query organizational catalogs and patch running graphs via `wf_graph_patch`
- Human-in-the-loop real-time graph modification and pause/checkpoint resume support
- Dual operational modes: DAG Workflow Orchestration Mode and Headless OpenAI-compatible API Service Mode
- Built-in database retrieval supporting local SQLite with vector search (bge-small-zh-v1.5/BM25) and MySQL/PostgreSQL
- Cron-like scheduled workflow runs with peak/off-peak budget-friendly execution policies
- Granular per-node configuration for system prompts, models, thinking intensity, tool presets, and ReAct loop bounds
Useful For
- Orchestrating multi-step AI pipelines such as automated content production, research, and review workflows
- Deploying DSH agent workflows as independent headless REST APIs for external integration with bots or custom frontends
- Designing and debugging complex autonomous agent teams with real-time visual tracking and human intervention
Who It Fits
- AI engineers and developers building complex multi-agent systems in DeepSeek Harness
- Teams needing visual workflow pipelines with organizational memory and continuous self-evolution
- Developers deploying autonomous DSH workflows as standalone API microservices
Documented Limitations
- Specifically tailored for DeepSeek Harness 0.2.0-rc.2 and not compatible with Alpha versions
- Requires specific build permissions (allowBuilds) in pnpm-workspace.yaml for packages like onnxruntime-node and sharp
- Initial agent meta-orchestration performance depends on external prompt quality during cold start
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, 13 development, 2 peer, and 2 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.
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
- GZX2211/dsh-Visual-Workflow
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- 7df3463 · 2026-10-16
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
- 49GitHub stars
- Forks
- 6
- Open issues
- 5
- Last commit
- 2026-10-11
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Persistent multi-model agent teams with dynamic DAG orchestration, run monitoring, and token insights for DeepSeek Harness.
View plugin →Native SQLite-backed task management and project board bundle with Web UI and Agent workflow tools for DeepSeek Harness.
View plugin →Comprehensive long-term memory, self-evolution, skills, task management, and multi-agent dispatch bundle for DeepSeek Harness.
View plugin →Second-model AI auto-review for DeepSeek Harness approval requests with fail-closed fallback and session audit logging.
View plugin →