terminal-ops

Executes repo commands, debugs CI failures, and reports verified git and test state.

Updated Mar 18, 2026
One-click install
npx skills add https://github.com/freedom909/real-estate-saas --skill terminal-ops-freedom909
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: terminal-ops
Source: https://github.com/freedom909/real-estate-saas/tree/main/.trae/skills/terminal-ops
Command: npx skills add https://github.com/freedom909/real-estate-saas --skill terminal-ops-freedom909

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When you ask an AI to fix a bug, run a command, or debug a CI failure, vague answers like "it should work now" are useless. This Skill enforces an evidence-first terminal workflow so every claim about what was run, changed, verified, committed, or pushed is backed by actual command output and git state. ## Core Features & Use Cases - Evidence-First Execution: Inspects errors, files, and git state before editing, then reruns the proving command before claiming anything is fixed. - Narrow Fix Discipline: Solves one dominant failure at a time with the smallest useful proving command, avoiding repo-wide churn and destructive git operations. - Exact Status Reporting: Distinguishes inspected, changed locally, verified locally, committed, pushed, and blocked so you always know the true state of your branch. - Use Case: Your CI pipeline fails on a pull request. The Skill reads the failing log, inspects the relevant test, applies a minimal fix, reruns the exact failing command to verify, and reports the branch and push result precisely. ## Quick Start Ask the AI to debug the failing CI test in this repo, apply a narrow fix, verify it locally, and push the branch with exact proof of what ran.

Frequently Asked Questions about terminal-ops

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

FAQPage Schema
How do I debug a CI failure with an AI assistant?▼

Point the AI at the failing CI log and repo path, then have it inspect the error and relevant test before editing. It should apply a narrow fix, rerun the exact failing command locally, and report verified status before any push.

How to verify a code fix actually works before pushing?▼

Rerun the smallest proving command that originally failed, such as the specific failing test, after applying the fix. Only escalate to a broader build or test pass once the local failure signature is resolved.

Can an AI safely run git commands on my repository?▼

Yes, when constrained by guardrails: inspect git state before editing, avoid destructive commands, respect unrelated local work, and only claim pushed when the branch actually moved upstream with the target branch named.

Why does my AI claim code is fixed when tests still fail?▼

This happens when the model answers from stale memory instead of live repo state. An evidence-first workflow prevents it by requiring the proving command to be rerun and its output cited before any fixed or verified claim.

When should I not use an evidence-first terminal workflow?▼

Avoid it for open-ended design work, greenfield feature planning, or broad refactors where no single failing command defines success. It is built for narrow, verifiable execution tasks, not exploratory coding.