🔁工作流与自动化Workflow & Automation 37
-
★ 2
940842546/dsh-permissions
Claude Code-style permission rules engine: hard/deny/ask/allow tiers with a hard tier above full access, workspace-scoped rules, wildcard path protection, and a visual staged editor; rules persist in settings.yaml.
-
★ 10
940842546/dsh-usage-billing
Usage and cost statistics with peak/off-peak pricing after the 2026-08-17 rate change, a main-UI summary panel, per-session breakdowns, and a usage heatmap.
-
★ 33
biociao/dsh-science
Claude Science-style research workbench: ReAct research-loop engine (research_* tools), versioned artifacts with provenance (artifact_* tools), and 10 science skills for genomics/pathogens/bioinformatics.
-
★ 18
btspoony/dsh-advisor
Pair a second model that passively reviews each turn and injects notes.
-
★ 58
btspoony/mstar-harness
Skill-driven harness/loop engineering workflow agent plugin.
-
★ 0
chenzhi-clude/dsh-hooks-pack
One-click Claude Code & Codex hooks: auto-discovers your existing ~/.claude / ~/.codex hook config and runs it on the official bridge plugins, with an explicit sandbox-policy shim so hook commands never die silently.
-
★ —
dsh-plan-execute
Dual-model plan/execute routing: a planner model thinks, an executor model acts.
-
★ 2
EvilIrving/dsh-proof
Independent read-only acceptance layer: spawns a read-only verifier before each top-level turn closes and steers non-pass gaps back into the agent.
-
★ 7
fakechris/dsh-track
Embedded task management engine: decision-point protocol, idea capture wall, Linear-style issue store.
-
★ 16
fuhefei/dsh-sentinel
Condition-driven wakeup: durable file/command/http/process/webhook watches that wake the agent.
-
★ 3
huchunlinnk/deepseek-desk-rsi
Recursive Self-Improvement engine: bounded perceive→integrate→verify→parity→repair→propose loop + 128-plugin 1:1 parity gate — DSH maintains DSH.
-
★ 116
icetomoyo/dsh_workflow
UltraCode-style multi-agent orchestration: a generatable, savable, governable, observable, resumable workflow layer.
-
★ 1
Jesse-njx/dsh-routines
Scheduled agents on a cron: run a prompt on a schedule and get the digest where you already are, with overlap/missed-run/timeout safety defaults.
-
★ 1
file-planning
trailmap: disk-persisted execution planning — milestone state machines, dependency tagging, audit events, retrospective notes.
-
★ —
task-board
Cross-session event-sourced work ledger: task tracking, audit history, kanban export.
-
★ 1
dsh-product-subagent-console
A conversation-level multi-agent workbench for editable task plans, real child-session trees, plan-to-runtime comparison, and evidence-backed recovery previews.
-
★ 8
Letter2025/dsh-approval-llm
Model-based permission approval: an approval-request answerer backed by a separate reviewer model.
-
★ 3
Letter2025/dsh-model-failover
Two-level model circuit breaker with failover: trip a model or a whole provider after repeated request failures and route the next request to a configured fallback.
-
★ 4
lonelymoon87/dsh-specflow
Adds specification artifacts, skills, commands, goal-backed implementation, and task-progress context.
-
★ 1,417
NanmiCoder/dsh-agent-teams
AgentTeams multi-agent teams.
-
★ 21
omdsh-dev/dsh-deep-research
Adaptive deep-research orchestrator built on the official workflow engine.
-
★ 7
omdsh-dev/dsh-inspect
Adversarial checkup → fix → review loop toolset.
-
★ —
odai-dsh-plugin
Profile-wide DSH governance and routing with a Web Control Center for responsibility and evidence inspection, plus compaction, scoped semantic memory, safety continuity, and verified delivery; compatible with DSH 0.1.5-rc.1.
-
★ 19
PerryLink/dsh-doublecheck
Engineering-discipline guard: grill the requirements before the first edit, enforce red/green test evidence gates, and audit the delivery with a forked adversary (grill-requirements skill + tool-policy gates).
-
★ 6
PerryLink/dsh-github
Production-grade GitHub CI integration: a composite action, a polling PR review bot (idempotent inline comments), and a status-check gate; all writes pass through approval.
-
★ 10
PerryLink/dsh-local-ai
Local Ollama model integration: discover/pull/remove/inspect, route by task type or keyword with automatic cloud fallback, and a /ollama overview.
-
★ 0
PerryLink/dsh-team-rooms
Persistent, cross-session multi-agent team rooms for DeepSeek Harness — members, a message bus, a shared task board, and a shared timeline that survive restarts.
-
★ 5
PerryLink/dsh-translate
Parameter translation across 11 providers (temperature, top_p, etc.) plus deterministic JSON repair; never fabricates data.
-
★ 3
Sev7een/dsh-plugin-automations
Settings-based scheduled tasks that run on time or during DeepSeek off-peak hours, with one-time and daily schedules backed by durable task state.
-
★ 7
timwhitez/dsh-self-evolving
Evidence-first, crash-resumable self-evolution engine for DSH: bounded Cordis candidate generation, one-shot real-Loader admission, Harbor evaluation, and an auditable journaled lineage.
-
★ 86
titanwings/dsh-automation
Scheduled coding runs in fresh agent sessions with auditable history.
-
★ 11
titanwings/dsh-plannotator
Plan review with anchored annotations and structured feedback back to the agent.
-
★ 6
vlln/dsh-loop
Recurring loops: `/loop` command + loop tool + activity status bar.
-
★ 1
weibaohui/dsh-process
Process management: brings ntd-style processes (multi-stage, multi-step agent workflow templates) into the dsh web UI - browse, edit, validate, import/export and AI-generate processes; the built-in library is read-only while the personal library is writable with live file sync, and agents read the library through process_* tools and advance work stage by stage.
-
★ 0
weibaohui/dsh-tasks
Scheduled tasks: run a prompt on cron schedules, each run opens a new agent session to do the work, with workspace binding, manual run, session auto-naming, and a fullscreen task management page.
-
★ 0
weibaohui/experts-management
Expert manager: manage ntd-format experts and expert teams (plugin.json + Agent MD + skill sets), browse and install from a built-in market of 50+ experts, and run tasks in an expert persona via /expert-<name> without consuming model directory tokens.
-
★ 2
whateverboy2333/dsh-flat-teams
Leaderless flat agent teams: cross-window structured task dispatch (state machine + event stream + offline resume), a read-only progress recorder, and a two-level web dashboard.
No plugins match.
Installing
# from npm (prebuilt) dsh plugin --profile web add <npm-package> # from GitHub (first run asks for allowBuilds approval — follow the hint, retry) dsh plugin --profile web add github:<owner>/<repo>
GitHub-sourced plugins run build scripts on your machine at install time. Only install sources you trust, and pin a commit (github:owner/repo#sha).
Get your plugin listed
Open a PR against awesome-dsh-plugin — one line each in every README language file; the site syncs automatically. Add the dsh-plugin topic to your repo too.