dsh-swarm-orchestrator
Role-based AI swarm orchestration plugin for DeepSeek Harness with model pinning, task DAG execution, and a live web dashboard.
Install
$ dsh plugin --profile web add github:linkbag/dsh-swarm-orchestratorPlugin 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
- Role-based AI swarm orchestration with architect review, parallel builders, automated/human reviewers, and integrators
- Per-role model pinning with fallback chains, reasoning effort ladder adjustments, and live catalog integration
- Parallel task DAG execution with dependency tracking (`blockedBy`), write scopes, and machine-checked evidence contracts
- Live Swarm web dashboard with Board (Kanban), Flow (DAG visualizer), and Roster (duty table and runtime config) views
- Robust failure handling including adaptive concurrency, watchdog reclaim, circuit breakers, and exponential backoff retries
- In-process memory-safe execution with configurable global and per-run concurrency limits
Useful For
- Orchestrating multi-agent research, data curation, and code refactoring workflows in parallel
- Enforcing automated multi-step verification and reviewer quality gates before marking tasks complete
- Routing specific swarm sub-tasks (e.g., architecture, code generation, review) to specialized LLM models
- Visualizing and managing complex multi-agent DAG pipelines inside the DeepSeek Harness web interface
Who It Fits
- Developers building multi-agent workflows and complex DAG tasks within DeepSeek Harness
- Teams requiring supervised AI pipelines with automated review loops and human-in-the-loop approvals
- Power users seeking granular control over per-role model pinning and concurrency limits
Documented Limitations
- Task agents execute in-process sharing the host Node.js heap; high concurrency across multiple swarms requires careful memory tuning
- Member messaging is one-directional; agents report progress to the dispatcher rather than communicating directly with each other
- Concurrent processes editing the same workspace swarm state are not supported; requires one host per workspace
- Evidence contracts are strictly enforced for command execution, but file evidence checks act as advisory warnings
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 0 runtime, 21 development, 9 peer, and 0 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.
package.json declares a prepare lifecycle script that may run during relevant package installation workflows.
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
- linkbag/dsh-swarm-orchestrator
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- b19b9c6 · 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
- 9GitHub stars
- Forks
- 1
- Open issues
- 1
- Last commit
- 2026-10-18
- 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 →Ontology-first research preset and epistemic loop plugin for DeepSeek Harness.
View plugin →Second-model AI auto-review for DeepSeek Harness approval requests with fail-closed security.
View plugin →