Plugins

Persistent provider-reported token usage totals with per-session, per-model, and daily activity statistics.

Optimize Prompt button under the composer: one click rewrites your draft into a clearer, more executable prompt, with a before-and-after dialog and one-click replacement.

DeepSeek Harness plugin: every user prompt you type (in the web UI) is appended to your atuin shell history, so it shows up in atuin search and shell integration.

A DeepSeek Harness plugin that turns the anywhere-claude-mem git sync into a DSH peer: pull cross-machine memory at session start and push new prompts/observations as they land in the claude-mem worker

Reusable Markdown prompt profiles for DeepSeek Harness with per-turn model selection, argument substitution, and state restoration.

Native OS notifications for approval and question/plan prompts: the toast names the workspace, click jumps to it, bilingual (zh-CN/zh-TW/en), with chime.

Scans skills and MCP configs for prompt injection, homoglyphs, hidden Unicode, dangerous shell, and credential leaks.

Infix rerun for DeepSeek Harness: re-run any earlier turn in place - the reply is regenerated from the same prompt while every later turn survives, so subsequent model calls read A B C1 D E F G. Implemented with the official surface-replace contract plus

Uses a configurable system-prompt section to request a chosen reasoning language (Simplified Chinese by default), and expands thinking rows by default while preserving rows the user manually collapses.

Configurable Git commit and pull request prompt instructions for DeepSeek Harness

Third-party DeepSeek Harness plugin: per-workspace custom prompt configured from a slash command and injected into the model context at session start.

cc-switch-style global prompt for DeepSeek Harness (dsh 0.1.5): one user-editable block injected into the system prompt of every conversation, mirrored into each session project's AGENTS.md, and naming the harness's user-global AGENTS.md by absolute path

Read-only side branch in the right sidebar: quote text to ask follow-ups without polluting the model context of the main session; every follow-up rebuilds the branch from the latest settled turns of the main session, and writes are refused at the tool-execution layer so the prompt prefix stays cacheable.

Progressive long-term memory for DSH: sharded markdown bodies under ~/.dsh/memory/topics/, with only a live-generated index (shard · count · path · per-entry line numbers) injected into the system prompt — so the prompt cost stays flat however large the m

Run the same prompt through every model you have, and see the difference without squinting.

Every change to your agent's instructions is recorded with a diff, and any change can be undone.

In-place prompt editing in the web chat: cleanly forks session history at the edited turn, preserves image and file attachments, and auto-archives the superseded branch.

Registers the Zemingxingxiao five-element brand-diagnosis skill with five companion reference skills (theory, diagnosis playbook, 240-case library, visual templates, glossary), a system-prompt router, and the mandatory HTML report template with 13 theory diagrams.

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.

dsh拓展全局提示词插件

Session-local scheduled tasks: inject a prompt into the current session after a delay, at a fixed time, or on a repeating interval, so the model wakes up and runs it on time. Manage them from a panel above the composer, the /schedule command, or seven model-facing tools. Timers freeze while DSH is closed and never auto-resume; pause keeps the remaining time, so resuming continues from where it stopped.

Maintained fork of dsh-approval-gate. A neutral operation whose confirmation count has reached the threshold auto-approves when the judge is unavailable, instead of prompting a human again; approver-facing explanations are generated in Chinese from the real sandbox mode, command and paths. Also carries deterministic hard-deny for credential exfiltration and system-path destruction, judge-input redaction, a judge model candidate chain, fingerprint-scoped allow rules, and an approval view with unified diff and one-click revert.

Long-term memory for DeepSeek Harness, read-only as of 0.1.0: a scribe_recall tool that reads a plain-markdown memory room, path rules that refuse directory traversal, Unicode smuggling and NTFS alternate data streams, and a bounded index whose truncation is always reported rather than silent.

Rapid-refill circuit breaker for DeepSeek Harness automatic compaction. It refuses a compaction that would be futile — the context refilled within fewer than N tool turns M times in a row, which the unbounded step-pressure trigger cannot detect — and instead of burning a summarization call on every step it reports the situation with advice, usually that a single read or tool output was too large. Shows state and re-arms through /compaction-breaker, and injects one prompt section so the model relays it to the user. Covers host-plane sessions through its profile bundle; ordinary sessions need one row replaced inside an agent preset.