dsh-ha-orchestrator
Model high-availability failover + subagent orchestration (fanout/pipeline/supervisor) for DeepSeek Harness
13 results
Model high-availability failover + subagent orchestration (fanout/pipeline/supervisor) for DeepSeek Harness
Same-runtime cross-Session discovery and communication for DeepSeek Harness
DeepSeek Harness multi-agent orchestration plugin with DAG scheduling, supervisor planning, bounded recovery, and runtime diagnostics for Cordis.
Add a Restart DSH button to the DeepSeek Harness WebUI. The button stops the dsh web process; an external supervisor (e.g. macOS launchd LaunchAgent) restarts it automatically so the UI comes back.
Always-on supervisor agent for DeepSeek Harness: the main-agent preset plus the schedule overlay, installable as one dsh bundle
项目 Agent 管理: a supervisor (主管) workspace + N dev-project workspaces, with logical isolation, task handoff, and a sidebar panel for DeepSeek-Harness agents.
Supervisor daemon for DSH Web resilience — auto-detect crashes, rollback to LKG, report
Workspace-owned durable process supervision, readiness checks, and a Runtime Center for DeepSeek Harness.
Watchdog for the dsh web GUI: crash logging + exponential-backoff auto-restart supervisor + status panel
Drain-aware supervised restart, Web status countdown, and same-session continuation for DeepSeek Harness.
A durable, bounded lifecycle supervisor with scheduled evaluation for live DeepSeek Harness sessions
Out-of-tree DSH plugin: canonical per-workspace Cordis scopes (workspaceCordis) with optional <workspace>/.dsh/cordis.yml composition mounting and live reload on top-level config changes; rc.6-compatible MCP core (transport/tools/connection supervisor) under ./mcp.
Self-management & self-evolution for DeepSeek Harness: health snapshots, config pre-check, blue-green evolve with auto-rollback, and an on-demand standby supervisor — let dsh manage itself.