为下一个工作流
找到对的 DSH 插件

Task planning with experience muscle-memory: condition-reflex recall of past solutions, LLM capability matching, and auto-persisted lessons.

UltraCode-style multi-agent orchestration: a generatable, savable, governable, observable, resumable workflow layer.

GPT plans dependency DAGs, DeepSeek workers execute tasks with adaptive parallelism, and an optional GPT reviewer audits incremental Git diffs and drives repair rounds.

AgentTeams multi-agent teams.

Reusable agent squads with per-agent provider/model routes and tool policies, serial/parallel dispatch, spawn/fork/chain context modes, and a Web management panel.

Scheduled coding runs in fresh agent sessions with auditable history.

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.

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.

Plan review with anchored annotations and structured feedback back to the agent.

Recurring loops: `/loop` command + loop tool + activity status bar.

Condition-driven wakeup: durable file/command/http/process/webhook watches that wake the agent.

Adaptive deep-research orchestrator built on the official workflow engine.

Adversarial checkup → fix → review loop toolset.

Embedded task management engine: decision-point protocol, idea capture wall, Linear-style issue store.

Pair a second model that passively reviews each turn and injects notes.

Adds specification artifacts, skills, commands, goal-backed implementation, and task-progress context.

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.

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.

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).

Skill-driven harness/loop engineering workflow agent plugin.

Evidence-first operating kit: six read-only tools (capability catalog, staged workflow plans, packaged skill reader, bounded local memory search, repository release audit, release checklist) and five packaged skills - plans and audits, never remote writes.

Model-based permission approval: an approval-request answerer backed by a separate reviewer model.

Adds an Auto Approve permission preset to the Web UI, using a fresh restricted Reviewer Agent to allow or deny each approval request.

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.