Plugins

Continues numbered and bulleted lists in the DSH composer, so Shift+Enter after "1. item" starts a new line with the next marker.

Shows the DeepSeek account balance at the bottom of the sidebar, with a configuration card on the Plugins page.

AI Look-Up: detects aimless grinding (repeated commands, failure loops, long runs with no output) and injects mid-run self-review nudges, plus a real local-clock anchor so the agent stops estimating elapsed time.

将 WorkBuddy 桌面 App 包含的模型自动接入 DeepSeek Harness — bring WorkBuddy desktop app models into DeepSeek Harness with zero configuration.

Universal Model Context Protocol (MCP) bridge for DeepSeek Harness: connects agents to stdio/sse/websocket MCP servers with a 10k dormant scale engine, sub-millisecond BM25 tool search, dynamic token budgeting, and a real-time web UI dashboard.

TypeSafe Jev as a first-class Cordis service and three model-visible tools for DeepSeek Harness — offline by default, egress disclosed, never default-on.

One-line clock context for DeepSeek Harness: a Current time timestamp reading (weekday, date, time, UTC offset, IANA timezone) opens every conversation turn and a Turn ended reading closes it, so the model always knows the time, date and idle gap across turns without asking.

Persistent Endeavour and Challenger peer sessions for DeepSeek Harness that plan, execute sequential tasks, and verify reported results.

Shows which graft index a dsh session is serving and whether it still matches the code, and gives the model graft_ask, graft_grep, graft_callers, graft_skeleton and graft_map as native tools.

Replaces the terminal tool on Windows with Git Bash.

Adds a reasoning-level editor with OpenAI and Anthropic protocol presets (llm-pi-ai), input-modality switches and a 1M/128K context and max-output preset to model rows in Settings.

Adds a changed-file summary row to finished turns and reverts or reapplies the file changes of the latest turn in the session.

Recalls prompts sent earlier in the current session with the up and down keys in an empty composer, and shows a position badge while browsing.

Governed long-term memory for DeepSeek Harness: append-only evidence ledger with write guards, 7-tier authority claims, supersede chains, and budgeted pre-step injection with per-domain quotas.

Research-debt guard for DeepSeek Harness: nudges coding agents to search docs, GitHub and the web before repeated local trial-and-error.

Let your AI agent see and operate a real Android phone: phone_look (vision + UI-tree fusion), tap/swipe/type, screenshot — over adb, for any MCP client.

Forgetting-curve in-place dialogue compression for DeepSeek Harness: recent turns stay verbatim, middle turns are structurally dehydrated (zero-LLM), ancient turns fold into a bounded summary pair — unbounded conversations on a bounded memory budget.

Conversational Markdown mindmap workspace: edit complete documents step by step, keep a live right-side visualization synchronized, and open, manage, and export mindmaps from the working directory.

Superself's self CLI as model-facing tools (project context, work units, reports, decisions) plus a /self command — cross-session, cross-project project state kept outside the code repo.

Evidence-driven company and job due-diligence for DSH: logs sources with evidence levels, adds evidence-bounded claims, verifies the legal entity against official registries, and renders a PROCEED/VERIFY/STOP report with a verification checklist and interview questions.

Complete mobile PWA for DSH, built on dsh-mobile-gate: secure remote-access gateway plus install-to-homescreen (manifest + service worker), offline capability, touch gestures (pull-to-refresh, edge-swipe back, pinch-zoom font), agent-done push notifications, and touch-first layout — desktop unaffected.

PR evaluation and improvement consultant: assesses PR success probability based on historical anti-patterns and success patterns, with one-click scoring and actionable suggestions.

Agent-initiated voice calls: `offer_call` rings the human (接听/拒接/稍后再说); accepted calls synthesize and play locally via CrispASR + Qwen3-TTS (9 speakers, 2 Chinese dialects), rejected calls return the decision to the agent.