dsh-autogate
DeepSeek Harness 自动审批插件:在 workspace-write 沙箱之上增加 Auto 档——确定性规则放行/拒绝,模糊操作由 LLM 裁决,危险操作转人工。保留工作区沙箱边界,不放宽为 full-access。
91 results
DeepSeek Harness 自动审批插件:在 workspace-write 沙箱之上增加 Auto 档——确定性规则放行/拒绝,模糊操作由 LLM 裁决,危险操作转人工。保留工作区沙箱边界,不放宽为 full-access。
DSH web plugin: adds a one-click Compact button and a New Session button to the conversation context meter panel. Compact submits /compact to summarize older history; New Session starts a fresh session in the same workspace (agent preset and permission settings follow the deployment default).
Academic writing toolkit for DeepSeek Harness agents: paper outlines, title/abstract skeletons, GB/T 7714 / APA / MLA citations and reference lists, LaTeX snippets, structure checks, term glossaries, phrasing QA, and a pre-submission checklist
Clinical documentation toolkit for DeepSeek Harness: SBAR and I-PASS handovers, SOAP and APSO notes, admission H&P, discharge summaries with pending-results enforcement, patient-instruction readability scoring, HIPAA Safe Harbor de-identification, and cited FDA label/FAERS/recalls/Drugs@FDA, MedlinePlus, PubMed/PMC/Europe PMC, and ClinicalTrials.gov (search/details/eligibility/investigators/endpoints) lookups. Drafts only; not a medical device.
Agent-decided approvals for DeepSeek Harness: a workspace-write base permission mode where an independent approval subagent judges every sandbox escalation (risky operations are rejected), with a configurable approval model and a per-session audit trail in the conversation window's 审批 tab.
A middle permission tier for DeepSeek Harness between workspace-write and danger-full-access: auto-approves harmless commands and operations targeting configured trusted areas, beyond the current workspace.
Collapse very long composer drafts into a summary card without changing DSH message submission.
DeepSeek Harness long-run mission suite: mission control, Long-Run Captain preset, patched LLM-as-a-Verifier, self-wake timer.
自然语言查询 + 商业智能报表插件 for DeepSeek Harness:在 dsh-mysql 连接底座之上,提供 Text2SQL 自然语言查库、15+ 图表类型、Dashboard 仪表盘、指标/维度管理、多维分析、自助分析、报表导出、审计日志与数据权限。Natural-language data queries + enterprise BI reports for DeepSeek Harness: Text2SQL, 15+ chart types, dashboards, metric management, multi-dimensional analysis, self-service analysis, export, audit log and data permissions.
DeepSeek Harness plugin: real-time Feishu/Lark notifications when a DSH session pauses for user interaction (permission asks, questions, fatal errors).
DeepSeek Harness 工作区管理工具集:查看、创建注册、重命名、删除工作区记录(纯注册,不触碰磁盘目录);破坏性操作(删除注册、改注册路径)走用户审批 fail-closed。
Workspace Only switch for DeepSeek Harness: confine file access of every sandbox mode to the session workspace
Standalone DeepSeek Harness access-mode plugin with Default, No Edit, and Auto modes
DeepSeek Harness plugin: Lord/Serf agent protocol 0.5 skills (file-based multi-agent orchestration with bounded execution and schema-validated submissions)
DSH plugin: strip GPT-family tool-call sandbox_permissions that are not strictly wider than the current session.
Live Mission Control for DeepSeek Harness agents, tools, subagents, and token usage
DSH Auto Guard: a Claude Code Auto Mode-like command approval mechanism that adds an LLM safety net on top of full access, with rules, caches, file tracker, and sensitive path gates.
dsh plugins ported from eco-agent: permission gate, audit chain, memory tree
PRO-LONG-style programmatic memory for DeepSeek Harness: appends every session event to a per-workspace log.txt that the agent retrieves with grep/python. Built-in write probe, /prolong status command and permission-deny counter.
Persistent long-horizon autonomous mission runtime for DeepSeek Harness: dependency DAG, independent verification, replan-on-failure, lease ownership.
Autonomous (auto) mode permission classifier for DeepSeek Harness: a Claude-Code-auto-mode-like classifier over tools/pre-execute and approval/request, a selectable 'auto' permission preset, LLM semantic judge, git checkpointing, agent discipline guidance, and a web control page in Settings → Plugins.
One DSH Remote Host bundle: Noise projection, commands, prompt/stop admissions, pairing settings. Linux and Windows.
Strip sandbox_permissions / justification from model tool-call arguments so sandbox escalation is never triggered when the session already has sufficient permission.
DSH profile plugin: unconfined (GPU-capable) session sandbox plus per-operation user approval for writes outside the workspace or to protected files.