Plugins

Nine consolidated tools that give agents full access to the stool ops CLI: server management, database query/redis, log search/tail, CI/CD deploy and history, MFA codes, git operations, notes, todos, and misc utilities (accounting, weekly reports, audit, project listing, nginx presets).

Bundles the remote-shell skill for secure SSH, SFTP, Telnet, and WinRM remote operations, with an encrypted credential vault whose login and execution scripts never accept plaintext passwords.

Per-task isolated git worktrees with a full create/sync/finish lifecycle: task names auto-derive branches, merge targets are double-checked, and batch cleanup is built in — zero manual git operations.

ADB device & bench operations for DSH: device discovery, structured logcat (background streaming), apk install, file pull/push, and dumpsys performance snapshots.

Auto-monitor and reveal active file operations and changes in DeepSeek Harness (DSH) right sidebar

Cordis-native, evidence-driven full-stack engineering workflow: inspect-first planning, explicit verification evidence, and approval-aware gating for sensitive operations.

Tailscale remote-access operations panel for the dsh web GUI: health checks, HTTPS entry toggle via tailscale serve, macOS proxy bypass, relay server ops, and ACL snippet generation.

Custom script manager for DSH: turn verified operations into reusable scripts that run directly via /script or register as agent tools, with parameterized scripts, declared expected outcomes, and execution history.

DSH tools for operating a self-hosted Talebook library: search, download, upload and manage books and audiobooks, manage Legado book sources, and admin operations, with approval-gated writes.

Codex-style Git workbench: workspace-scoped changes, diff, stage, commit, guarded branch operations, graph history, fetch, fast-forward pull, and upstream push through official plugin APIs.

Auto-approval permission guard: a middle tier between workspace-write and danger-full-access — auto-allows safe operations inside trust directories, always asks a human for destructive ones, with 11 per-category switches and an audit trail.

Codex-style auto-review permission mode: adds an auto-review preset that auto-approves safe sandbox escalations, asks on risky or ambiguous ones, and rejects critical unconfirmed operations.

Cordis-native, evidence-driven full-stack engineering workflow: inspect-first planning, explicit verification evidence, and approval-aware gating for sensitive operations.

Safe Kubernetes capability for DeepSeek Harness with workload inspection, bounded logs and events, failure diagnosis, and approval-gated operations.

Connects DSH to local Beav for Xiaohongshu (RED/RedNote) and social-media AI operations: topic research, knowledge, copywriting, image, audio, and short-video workflows with approval and artifact tracking.

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.

Host-only command safety gate for DeepSeek Harness: blocks recognized dangerous operations and requests DSH approval when a command cannot be inspected.

GitHub operations for DeepSeek Harness: releases, tags, a guarded generic API pass-through, workflow runs, secrets and sanitized mirror publication.

Routes DSH filesystem, shell, subprocess, and PTY operations to local, SSH, and Docker execution nodes, with a node registry, an authentication gate, a mobile-aware management UI, and per-workspace custom project instructions.

Advanced sidebar operations for the DeepSeek Harness Web Client: git changes, multi-session terminals, a file browser, a dev-server preview, background tasks, Open in, Archive and Delete, in a resizable dock beside the conversation

Routes DSH filesystem, shell, subprocess, and PTY operations to local, SSH, and Docker execution nodes, with a node registry, an authentication gate, a mobile-aware management UI, and per-workspace custom project instructions.

Turns DSH permissions into 9 selectable presets (3 sandbox modes x 4 approval tiers), with rule plus LLM risk classification across four risk levels, password-gated approval for high-risk operations, and a JSONL audit log.

Runtime observability and plugin operations panel for DeepSeek Harness: shows the profile plugin tree, real Loader fiber state, capabilities/dependencies/compatibility, current-session plugin activity and upstream source/version checks, with one-click enable/disable/update/uninstall plus a standalone startup-failure diagnostics page that works while DSH is not running.

Live Kanban operations board for DeepSeek Harness Web