Convert task requests into dependency-aware execution plans and delegate to Manager AI.
npx skills add https://github.com/kinetas/harness_engineering --skill newtask-kinetas
Please help me install this Agent Skill. Skill: newTask Source: https://github.com/kinetas/harness_engineering/tree/main/newTask Command: npx skills add https://github.com/kinetas/harness_engineering --skill newtask-kinetas
It turns a natural-language task request into a structured, actionable plan that can be executed by an AI team with correct sequencing and tracking.
/newTask [description] into a categorized request type (bug, feature, improvement, refactor) and produces a clear summary.doc/AI_list.txt with the new Manager AI task slot, delegates execution, and then produces a completion acknowledgment while ensuring reporting flows to report/report.md.Invoke /newTask and include what you want changed or built, for example “Refactor the projectStart flow to reduce redundant steps and improve task dependency clarity.”
High-intent search queries and answers about installing and using this skill.
To turn a natural language request into a dependency-aware task plan, you can invoke the /newTask command with your project description. The system classifies the request, generates a task list with estimated impact areas and a dependency graph, and delegates execution to a Manager AI.
AI delegation works by converting your categorized request into an ordered task list with a dependency graph. It updates the AI_list.txt file with a new Manager AI task slot, delegates the execution in the correct sequence, and returns a completion acknowledgment.
Yes, you can use this for project execution scenarios like refactoring, improvements, bug fixes, and new feature development. It handles task sequencing and tracking across AI roles by generating a dependency-aware breakdown tailored to your specific project type.
You do not need to manually update state documents before generating an execution plan. The process reads system and state documents automatically, updates doc/company_state.json and doc/AI_list.txt during delegation, and ensures reporting flows to report/report.md.
The best way to monitor AI task execution is to rely on the built-in reporting flow. After the Manager AI completes the delegated tasks, the system produces a completion summary and automatically enables downstream monitoring by appending execution updates to report/report.md.
Task planning requires a dependency graph for AI orchestration to ensure work proceeds in the correct order. By mapping estimated impact areas and dependencies, the Manager AI can execute sequenced tasks accurately and prevent conflicts during complex project execution.
kinetasCreate Harness Engineering state files and initialize AI orchestration.
constellizeCollect memory, thread, throughput, and business metrics for long-running services.
AustinZ21Convert dependency-ordered spec-kit tasks into GitHub issues.
wayne930242Classify knowledge tasks and delegate them to appropriate Claude Code skills via the Skill tool.