Plugins

One standing system prompt for every workspace: a Custom instructions page in DSH's own Settings panel whose text is added to the system prompt of every session, subagents included, and stored in the profile instead of any project.

Per-request cost ledger covering retries and subagents, holiday-aware peak/off-peak pricing, shared pricing/usage/balance/budget services, and DeepSeek bill reconciliation.

Runs Superpowers as an agent preset and gives each skill its own subagent model from a Settings table, so TDD, debugging and verification no longer all inherit the parent model.

The child-scoped report tool for continuable in-process subagents on DeepSeek Harness: installs `report` plus its usage guidance into every continuable child scope, so findings reach the parent before the child finishes.

OpenRouter provider that reads the live /models list, so new models appear without a harness release: models sorted by name with input/output prices, per-model reasoning levels, image input, reasoning_details replay, Anthropic prompt caching, and actual spend per session including subagents from OpenRouter usage.cost, shown under the Web UI composer and by /cost.

Ships a skill about handing work to subagents instead of doing it inline - which tasks are worth fanning out, how many to run at once, how to send them in parallel, and how to check what they returned instead of trusting the summary.

Flip the DSH page over to a glass task board showing live jobs, subagents and session state, with an infinite canvas that unrolls a card into draggable controls, embedded web-app frames, and a conversation box that edits the canvas.

In-process DeepSeek Harness subagent provider that seeds a child with only the last N completed parent turns (fork-lite), leaving official fork untouched.

Unified management and invocation of external Agent CLIs in DSH, fully isolated from the user's native installs; preset Provider, model, reasoning effort, and permissions per CLI, callable directly or like native subagents.

Keeps every model request inside a per-provider allow-list, enforced at the agent/request waterfall: disallowed models are redirected to a fallback or refused; subagents stop inheriting stale creation-time routes; warns when consecutive requests would make a --models-max 1 llama.cpp router reload models. A one-entry list is a global single-model mode.

Explains which layer killed a failed turn: times every model request (first byte, longest silence, total) and combines that with the error to name the culprit (dsh idle watchdog vs Node undici timers vs SDK timer vs server crash, context overflow, gate queue, tool timeouts, failed compactions) and the exact setting to change; verdicts in the DSH terminal plus a /why command, subagents included.

Task Runner mode for DeepSeek Harness: the main agent decomposes large tasks into independent subtasks, dispatches them to subagents (default local model) in fresh short contexts, and synthesizes results; concurrency and worker/orchestrator models are configurable in the settings panel.

Auto-switch the DeepSeek route to the vision model on demand: flash main session switches (A), pro keeps deep reasoning and delegates image reading to a vision subagent (B), subagents always switch, with fatal-failure fallback. No manual model switching.

Specialist subagent delegation preset (orchestrator + 5 roles: oracle/designer/fixer/explorer/librarian) that seeds itself into your DSH home — background-first delegation, JSON-driven models/effort/tool permissions, scoped MCP for research.

Route fresh DeepSeek Harness subagents to registered provider/model pairs declared in role Markdown files

Lifecycle supervision, evidence-driven audit subagents, safe intervention, and blind acceptance gates for DeepSeek Harness

Adversarial code review — parallel finders inspect a diff through separate lenses (correctness, lifecycle, contract, security), then each finding goes to independent verifiers tasked with refuting it, and one refutation drops it.

Declarative task-DAG orchestration for DSH: validates dependency graphs, runs topological task layers in parallel through workflow-backed subagents, and propagates failures deterministically.