Execute LLM-generated DAG plans with dependency resolution and state persistence.
npx skills add https://github.com/t3ta/quedex --skill quedex
Please help me install this Agent Skill. Skill: quedex Source: https://github.com/t3ta/quedex/tree/main/.claude/skills/quedex Command: npx skills add https://github.com/t3ta/quedex --skill quedex
This Skill automates the execution of complex, multi-step tasks by managing their dependencies, parallelization, and state persistence, allowing LLMs to focus on planning rather than execution.
Create a plan file named 'my_plan.yaml' and then execute it using 'quedex run my_plan.yaml'.
High-intent search queries and answers about installing and using this skill.
To execute LLM-generated DAG plans with dependency resolution, define your workflow in a plan file like 'my_plan.yaml' and run it using 'quedex run'. The system automatically handles dependency resolution, parallel execution, and state persistence.
Automating parallel task execution for multi-step workflows is best handled using DAG-based plan files. You define tasks and their dependencies in a YAML plan file, and the execution engine manages concurrency control to run tasks in parallel safely.
Yes, you can use multiple backends like Codex CLI, Claude Code, and Opencode for task execution. The orchestration engine supports these backends to execute the individual steps defined within your DAG plan.
State persistence and recovery for interrupted workflow runs works by automatically saving execution state. If a run is interrupted or fails, the system resumes from the last completed step and recovers using the persisted state.
You monitor real-time progress of parallel DAG execution using the built-in real-time TUI and Web dashboard. These monitoring features track task statuses, retries, and overall workflow progress as it happens.
You should use DAG orchestration for software development workflows when automating complex sequences like researching a feature, implementing it, writing tests, and deploying, ensuring each dependent step completes successfully before the next begins.
WenyuChiouSplit a high-level goal into a DAG of subtasks and emit .coord/plan.yml plus per-agent task files.
kbanc85Launch the Brain Monitor TUI to display live neural pulse, identity, health, constellation, and landscape dashboards.
chronista-clubIntegrate ACP-enabled AI agents with Zed Editor via JSON-RPC over stdio.
jimmymalhanExecute tasks defined in a Directed Acyclic Graph with dependency validation.
