10x-implement

Implements approved technical plans phase-by-phase with progress tracking and verified commits.

Updated May 20, 2026
One-click install
npx skills add https://github.com/ZawilecxD/trAInR --skill 10x-implement-zawilecxd
Or copy as Structured Prompt for Agentβ–Ό
Please help me install this Agent Skill.
Skill: 10x-implement
Source: https://github.com/ZawilecxD/trAInR/tree/main/.cursor/skills/10x-implement
Command: npx skills add https://github.com/ZawilecxD/trAInR --skill 10x-implement-zawilecxd

SYSTEM DOCUMENTATION & REQUIREMENTS

πŸ’‘ This Skill includes references (resource) components.

What problem does it solve? Executing a multi-phase technical plan by hand is error-prone: steps get skipped, progress state gets lost between sessions, and commits mix unrelated changes. This Skill turns an approved plan.md into a disciplined, resumable implementation run with per-phase verification and clean Conventional-Commits history. ## Core Features & Use Cases - Phase-driven execution: Reads context/changes/<change-id>/plan.md, derives the next pending step from the canonical ## Progress section, and implements one phase at a time. - Verified phase-end commits: Runs automated success criteria, gates on manual confirmation, stages only touched files, and writes the closing commit SHA back into each completed Progress row. - Resumable state tracking: Progress lives only in the plan file β€” no sidecar state β€” so work can resume mid-plan or after a context reset using a generated resume command. - Use Case: After approving a plan for an OAuth login feature, run the Skill with the change-id to implement Phase 1, verify it, commit it with a Linear issue reference, and continue phase-by-phase until the change is marked implemented. ## Quick Start Ask the AI to implement the approved plan for your change by running /10x-implement with your change-id, for example the oauth-login change, and confirm each phase's manual verification as prompted.

Frequently Asked Questions about 10x-implement

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

FAQPage Schema
How do I implement a multi-phase technical plan with an AI assistant?β–Ό

Provide the change-id or plan path, and the Skill reads plan.md, finds the first unchecked Progress item, and implements phase by phase. Each phase ends with automated verification, a manual confirmation gate, and a Conventional-Commits commit.

How does the Skill track implementation progress across sessions?β–Ό

Progress lives only in the plan's ## Progress section as checkbox rows β€” no state files or markers. To resume, re-invoke with the change-id and phase number; the first unchecked row determines where work continues.

What happens when the code doesn't match the plan?β–Ό

The Skill stops, presents the expected versus actual situation, and asks how to proceed: adapt and continue, skip the part, or stop and re-plan. It never silently diverges from the approved plan.

Can it handle unrelated dirty files in the git working tree?β–Ό

Yes. Before each phase commit it compares git status against the touched-file set and asks whether to stage only planned files, stage everything, or abort. Commits always use explicit paths, never git add -A.

What are the limitations of plan-driven implementation?β–Ό

It requires a pre-existing approved plan.md with a Progress section and refuses archived changes. Empty-diff phases commit nothing and leave rows without SHAs, which surface as warnings during archival.