Find your next challenge.

The problem

Expose deterministic pre- and post-compaction integration points, including a post-compaction lifecycle signal, so plugins can restore or re-seed model-visible context after compaction without polling.

Possibly addressed by:Force Compact · Compactor · Manual Compact Plugin

The problem

Provide supported operations to rename workspaces and move sessions between workspaces, safely updating paths, metadata, storage, registrations, and caches without manual edits or restarts.

Possibly addressed by:Session Mover · Session Workspace · Session Manager

The problem

Allow session creation to select an agent preset through workspace/session APIs and headless CLI, including loading presets from .dsh/.agent-presets/ with an --agent-preset option.

Possibly addressed by:Preset · Workspace Presets · Agent Preset Switcher

The problem

Detect repeated, cyclic, or near-identical tool-call patterns across configurable windows and turn counts, then block or interrupt them and redirect the agent toward synthesis, clarification, or a different strategy.

Possibly addressed by:Loop Continue · Tool Failure Circuit Breaker · Repeat Stop

The problem

Add an independent vertical scroll area for the Settings dialog’s left navigation so all sections remain accessible in short or zoomed viewports. Keep the title fixed, preserve keyboard access, and prevent navigation scrolling from affecting the content pane.

Possibly addressed by:Settings Nav Scroll · Settings Sections · Settings Nav Organizer

The problem

Add bounded retries and graceful fallback from WebSocket to SSE or other HTTP transports, including compatibility with HTTP/2/HTTP/3 proxies and automatic reconnect with backoff. Preserve durable-stream safety by avoiding duplicate output or tool actions, and cover disconnect, cancellation, exhaustion, and fallback cases with regression tests.

The problem

Expose an additive, extensible UI slot immediately below the zero-turn new-conversation composer, preserving normal document flow, registration order, and session-specific visibility without replacing the resident input.

Possibly addressed by:Conversation Flat · Client Ui Selection Add · Subagent Panel

The problem

Let users navigate previously submitted prompts with Up/Down keys and recover from unexpected execution paths through a flexible rewind mechanism, across the relevant interfaces.

Possibly addressed by:Keyboard History · Input History · Prompt History

The problem

Allow UI and settings-plugin preferences to use host persistence for remote browsers that pass the trusted-host fence, including declared hosts, LAN names, and reverse proxies.

Possibly addressed by:Settings Remote · Trusted Host Proxy 403 Fix · Remote Settings Compat

The problem

Let users select or identify a file and line range and attach the relevant excerpt or review context directly to a message. Support consistent path-and-line references so agents can act on precise locations without requiring users to restate them.

Possibly addressed by:Improved Inline Edit · Inline Comments · Editor Selection

The problem

Let plugins rewrite tool-call arguments before execution and assistant text before display or persistence. Preserve the model-emitted arguments for faithful auditing while enabling redaction, secret restoration, or other safe transformations.

Possibly addressed by:Tool Rewind · Input Rewriter · Tool Args Unwrap

The problem

Provide a `dsh doctor` or `dsh diagnose` command that checks runtime versions, dependencies and native modules, Web UI prerequisites, and first-run configuration, then reports actionable fixes. This would simplify fresh-install troubleshooting and standardize diagnostic output for bug reports.

Possibly addressed by:Plugin Doctor · Version Update · Doctor

The problem

Provide built-in or reusable audio and browser notifications for important agent states, including waiting, blocked, and round completion, so users can monitor agents without continuously watching the harness.

Possibly addressed by:Chime Alerts · Notify · Agent Notify

The problem

Improve the Markdown math renderer so currency and other literal dollar signs are not paired with later delimiters, while preserving valid inline and display math. Add appropriate flanking and whitespace guards with literal-text fallback.

Possibly addressed by:Math Input

The problem

Implement layered, configurable safety controls that require explicit approval for risky, costly, broad, deployment-related, or harness-stopping actions. Include fencing and energy or spending limits to keep agents within mission scope.

Possibly addressed by:Agent Guard Dsh · Agent Approval · Danger Guard

The problem

Add an optional icon channel to `settings.section` registrations and propagate it through the ledger and renderer, so third-party plugin pages can display custom navigation icons instead of the fallback gear. The capability should be backward compatible and follow the existing `icon?: ReactNode` pattern.

Possibly addressed by:Ui Settings Icons · Settings Sections · Settings Nav Organizer

The problem

Let users package custom agent flows as reusable skills and define repeatable, on-demand tasks outside AGENTS.md. This enables workflows such as migrations or component generation without unnecessarily polluting the agent’s context.

Possibly addressed by:Agent Useful Skills · Skill Creator · Agent Skills

The problem

Notify users about pending questions, approvals, completed replies, and task errors when the DSH tab is hidden, with title flashing and configurable system notifications. Where supported, let users answer or approve actions directly from interactive notification cards, while suppressing notifications for the active foreground tab.

Possibly addressed by:Taskdone Notify · Web Notify · Approval Notify

The problem

Provide orchestration for multiple coding-agent teams or workers running in parallel, with shared relevant context, coordination, and ongoing feedback between them.

Possibly addressed by:Subagent Profile · Background Agents · Parallel Agent Ade

The problem

Allow users to control verbosity and clarity independently of reasoning, including reliably reducing unnecessary detail and expanding or explaining unfamiliar acronyms while preserving useful content.

Possibly addressed by:Output Style · Prompt Polish · Output Styles