devlyn:auto-resolve

Run a multi-phase build, evaluate, fix, and document pipeline automatically.

4|Updated Feb 23, 2026
One-click install
npx skills add https://github.com/fysoul17/one-query-v1 --skill devlyn-auto-resolve
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: devlyn:auto-resolve
Source: https://github.com/fysoul17/one-query-v1/tree/main/onequery/.claude/skills/devlyn%3Aauto-resolve
Command: npx skills add https://github.com/fysoul17/one-query-v1 --skill devlyn-auto-resolve

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Fully automated build-evaluate-polish pipeline. One command runs the entire lifecycle of implementation tasks—from planning to polished delivery—without human intervention, coordinating build, evaluation, fixes, simplification, review, and documentation across phases.

Core Features & Use Cases

  • End-to-end automation: orchestrates build, evaluation, fix loops, documentation, and governance in a single command.
  • Multi-phase collaboration: spawns phase-specific subagents and manages file handoffs to ensure deterministic results.
  • Cross-model orchestration: supports auto routing to Claude or Codex per phase with optional security/quality checks.
  • Audit-ready outputs: generates reproducible reports and changelogs for audit and handoff scenarios.

Quick Start

Describe your task and invoke devlyn:auto-resolve to run the full end-to-end pipeline automatically.

Frequently Asked Questions about devlyn:auto-resolve

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

FAQPage Schema
How do I automate an end-to-end software delivery pipeline without human intervention?▼

Automating an end-to-end software delivery pipeline without human intervention requires orchestrating build, evaluation, fix, and documentation phases. This skill runs a hands-free, multi-phase workflow that automatically builds, evaluates, fixes, simplifies, reviews, and documents implementation tasks from start to finish.

What is an autonomous multi-phase build and evaluate pipeline?▼

An autonomous multi-phase build and evaluate pipeline is an automated process that spawns phase-specific subagents to handle building, evaluation, and fixes. It orchestrates these phases sequentially, managing file handoffs to ensure deterministic, reproducible results across the entire software delivery lifecycle.

Can I route specific pipeline phases to different models like Claude or Codex?▼

Routing specific pipeline phases to different models like Claude or Codex is supported through cross-model orchestration. The pipeline can auto-route individual phases to the preferred model, enabling optional security and quality checks during the automated build and evaluation process.

How do I generate reproducible audit reports for automated code changes?▼

Generating reproducible audit reports for automated code changes is handled automatically during the pipeline's documentation phase. The workflow produces audit-ready outputs, including detailed changelogs and reports, ensuring reproducible results for audit and handoff scenarios after the build and review phases complete.

Does the auto-resolve pipeline work for continuous integration and delivery tasks?▼

The auto-resolve pipeline works for continuous integration and delivery tasks by automating the full implementation lifecycle. It coordinates build, evaluation, fix loops, simplification, and governance in a single command, making it suitable for zero-human-intervention execution from task description to finished work.

What are the limitations of fully automating software delivery pipelines with autonomous agents?▼

Limitations of fully automating software delivery pipelines with autonomous agents include the reliance on phase-specific subagents for deterministic results and the need for cross-model orchestration setup. Complex edge cases may require manual intervention if the automated fix loops cannot resolve evaluation failures.