Plugins

Model console panel: switch the model for every conversation at once or one at a time, filter sessions by idle time, fold subagents under their parent session, and bulk archive or release sessions.

Composer tweaks for the DSH input box: a Markdown source/view toggle, Markdown rendering for user messages, one mixed file-and-folder picker backed by the macOS open panel, a GenUI prompt switch inside the plus menu, an open-in-new-window action on sidebar Session rows (needs the separate `dsh-popout-window` host plugin), and a collapsible subagent tree. Requires DSH 0.2.x.

Blocks the end of a turn while temporary files from that turn are still lying around: it registers a listener on the serial agent/turn-stopping hook and, if anything is left, hands the model the file list plus two ways out — delete them, or justify each one and move it somewhere deliberate. Candidates come from two narrow temp-file sources, plus a backup-retention channel that is on by default and flags excess .bak copies of files touched this turn. The first is the .agent-tmp/ convention directory under the session cwd: whatever is left in it is reported outright, because that directory exists only for throwaway files. The second is the paths this turn actually wrote, captured through the fs/write-intent, fs/edit-intent and fs/delete-intent hooks; those are reported only when they sit under that cwd (compared after realpath normalisation), fall outside .git, node_modules, out, dist, build, target, vendor, .venv and the Python bytecode cache directory, match a temp-file pattern that requires a separator after the prefix so checkout.js and benchmark.ts are not taken for scratch files, and the file is still present — which keeps junk already sitting in the workspace from being resurrected as a complaint about this turn. One gate per turn, subagent sessions skipped, nothing deleted unless the backup mode is explicitly set to prune, and the injected wording follows the host locale instead of assuming a Chinese one.

Agent preset that hands every task to a Claude Code subagent, so the session only writes the task brief, delegates and reports the result. Requires a locally installed, logged-in Claude Code.

Tells the parent agent when a subagent died from an upstream failure instead of finishing: it listens to agent/error on every agent, waits a short settle delay, and if the child is still not running delivers a followup turn to the parent carrying the child id, the failure and the options the parent has, with a cooldown and a per-child cap so a flapping provider cannot flood the parent.

Passive recorder that replays the main agent and every subagent on one timeline, flags repeated calls, loops, recurring errors and truncated output, and writes an AI cause-and-effect review on demand.

Agent Teams management for the DSH Web GUI: inspect the teammates of any session, delete a role together with its session (the session folder moves to a recoverable trash), reclaim the freed member slot on pre-0.2.2-rc.1 teams, keep deleted members out of the subagent picker and the team roster, and restore a read-only members panel in the session header on 0.2.2-rc.1+ — works with both the old and the new Agent Teams implementations without patching DSH.

Lists this session's runtime errors — turn errors, tool failures, command failures and subagent failures — as copyable cards in a conversation tab.

Select-all cleanup for archived DeepSeek Harness sessions: deletes the session log, projection cache, workspace accounting and archive marker, cascades subagent logs, and can also remove unreferenced attachments, the request-image cache, the workspace files a Session wrote and leftovers of already-deleted Sessions.

Lets you manually pick the model for a teammate (subagent): when it is stuck on a model it cannot use (say, out of quota), switch it to one that works and let it carry on — no rebuilding, and your own conversation and the global default model are untouched.

Writes one model and reasoning effort into every session at once from a control beside the composer model picker, since the shipped picker only changes the session you are in. Sessions that already match are skipped, subagent sessions follow their parent instead of being written, and the same selection can be pinned as the default for new sessions.

Estimated CNY API cost for each conversation turn and for the session token total, using DeepSeek peak/off-peak and per-model pricing; subagent costs are listed separately.

General-purpose chat agent on a parent agent + worker sub-agent split, for DSH Desktop only: a sidebar row opens each assistant's long-running conversation, where the parent chats, records schedule, lists, memos, logs, profile, persona and projects as plain files in that assistant's workspace folder, and hands other work to background workers through the bundled complete-subagent, picking each worker's model from the ones you enable and approving their write access. A heartbeat wakes it for reminders, a daily brief and wake-ups it sets itself, and holds back during quiet hours. Through the bundled im-robot it chats over WeChat: the model decides per reply with im_route whether the reply also goes to WeChat, and images and files sent from WeChat appear in the conversation and stay in the workspace's IM/WeChat folder for a set number of days. Its tools never appear in normal sessions.

File-based task pool and dispatcher for DeepSeek Harness: keeps a P0-P3 backlog as Markdown, hands one task at a time to a background subagent, gates delivery by rerunning the project test command, and accepts the result against a git commit or a file snapshot; git and non-git directories both work.

External one-shot subagent provider using the local CodeBuddy CLI over ACP, with background result delivery and steering of running children.

Adds lasting bots with their own name, photo, persona, memory, voice and model, a Bots panel in the sidebar, and a hidden helper subagent per bot that follows the conversation, searches ahead and leaves notes the bot reads; voice uses your own ElevenLabs, OpenAI, Groq, Mistral or Deepgram key. Interface and prompts in Brazilian Portuguese.

After user setup, guides native subagents toward model preferences by task type, using native model-selection permissions with no extra routing-model calls.

Requirement-to-evidence audit for DSH sessions: maps each requirement to tool calls, files and commands, records per-requirement verification, and sends missing items back to the agent.

Focus-aware notifications for DeepSeek Harness: approvals, questions, plan reviews and finished replies appear as one alert only when the conversation window loses focus; background conversations alert at any time and subagent sessions never do.

Two small macOS features in one plugin. keepAwake stops the display and system from sleeping while dsh runs (caffeinate tied to the host process). taskNotify posts a system notification when a top-level conversation finishes its turn, with the outcome (completed, error, blocked, output cap) and the duration; clicking the notification brings DeepSeek Harness forward. Click-to-focus needs Xcode Command Line Tools to build the notifier; without the toolchain the notification still appears but clicking it opens Script Editor. Subagents and Agent Teams members never notify, and a stop you pressed is silent by default. Both features can be switched off separately.

Soft-branches a self-contained sub-question into a child session and returns only a bounded digest, so the main context grows by a short summary instead of the whole investigation.

Cost display for AiSpot (sapi) relays: crawls the provider's group list from its public pricing endpoints instead of hardcoding it, lets you pick the active group from a dropdown in the conversation header, and shows a per-turn cost at the end of each turn plus a session total that recurses through every subagent subtree. Switching groups re-prices all history instantly, because the host publishes token buckets only and the browser applies the live rates.

Shows where every session's tokens go: lifetime totals, cache hit rate, peak/off-peak cost, a calendar heatmap, a trend chart, per-model and per-workspace breakdowns, and a session tree revealing subagent sessions, with CSV/JSON export. Also ships context clear/compact actions and a prompt optimization button.

Cross-project session grouping for the DSH sidebar: define named sections, assign sessions from any project into them, then collapse, rename, archive or move a whole project in one action; a toggle in the Sections panel swaps the host session browser for the section tree and restores it on demand, and each row flags live work — its own turn, a subagent it delegated to, or a background job still going — or a completion you have not looked at yet.