uexecute

Execute approved developer task plans into repository changes with phase dispatch and plan-diff verification.

47|12|Updated Apr 17, 2026
One-click install
npx skills add https://github.com/btseytlin/ultrapack --skill uexecute
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: uexecute
Source: https://github.com/btseytlin/ultrapack/tree/main/plugins/up/skills/uexecute
Command: npx skills add https://github.com/btseytlin/ultrapack --skill uexecute

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

uexecute turns an approved task plan into a working implementation while enforcing plan alignment, safe scope boundaries, and consistency across phases.

Core Features & Use Cases

  • Plan-driven execution: Dispatches each plan phase to an appropriate implementer subagent and ensures phases honor the approved plan.
  • Interface-graph wave scheduling: When a plan declares an interface graph, runs phases in correct topological waves while respecting blocking vs non-blocking dependencies.
  • Safety and correctness guardrails: Runs plan-diff checks and consistency sweeps after each commit, and forbids silent fallbacks or mutations of external specs.
  • Failure-aware workflow: Handles DONE_WITH_CONCERNS, NEEDS_CONTEXT, and BLOCKED states by reconciling deviations or re-dispatching with corrected context.

Quick Start

Use uexecute to implement the approved plan from docs/tasks/<slug>.md for the current task and commit phase-by-phase with plan-diff and consistency checks enabled.

Frequently Asked Questions about uexecute

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I execute a multi-phase developer task plan safely?

Interface-graph wave scheduling runs plan phases in correct topological waves, respecting blocking versus non-blocking dependencies to enable parallel implementation without violating execution order.

What is interface-graph wave scheduling for parallel code implementation?

Interface-graph wave scheduling runs plan phases in correct topological waves, respecting blocking versus non-blocking dependencies to enable parallel implementation without violating execution order.

How do I validate code consistency after committing a task plan phase?

You validate code consistency by running post-commit consistency sweeps and strict plan-diff checks after each phase commit to ensure repository changes align with the approved plan.

Does plan execution handle blocked states or missing context without silent fallbacks?

Plan execution handles BLOCKED, NEEDS_CONTEXT, and DONE_WITH_CONCERNS states by reconciling deviations or re-dispatching with corrected context, strictly forbidding silent fallbacks.

What are the limitations of automated plan execution regarding external spec inputs?

Automated plan execution requires read-only handling of external spec inputs and enforces wave disjointness checks on interface paths, preventing unauthorized mutations during phase orchestration.