Profile My skills Saved Collections Edit profile Submit a skill

← all publishers

tnagatomi

@tnagatomi source repo

3 published skills

  1. ▌
    Chunked Impl · tnagatomi
    Implement a feature or fix incrementally by splitting the work into small chunks and building each chunk test-first with the tdd skill's red-green-refactor loop, then committing once tests are green. Depends on the tdd skill (https://github.com/mattpocock/skills/tree/main/skills/engineering/tdd) for the per-chunk red-green-refactor discipline. Use this skill whenever the user asks to implement, build, or fix something with phrases like "implement X chunk by chunk", "commit as you go", "test and commit each step", "incremental implementation", "split this into small commits", or any time the user wants steady, verifiable progress with a clean git history rather than a single big change. Prefer this skill for non-trivial changes that have (or should have) automated tests.
    0
    installs
  2. ▌
    Agent Test With Report · tnagatomi bundle
    Drive the agent-browser CLI to exercise the features a pull request changes, then post an evidence-only findings report back to the PR. Depends on the agent-browser skill (https://github.com/vercel-labs/agent-browser) for the actual browser operation. Use this skill when the user wants an agent to manually test a PR in a real running app and leave a report on the PR, with phrases like "agent-test this PR", "browser-test the PR and comment results", "run exploratory testing and post a report", "check this change in the running app and report on the PR". The report collects evidence (operations, screenshots, findings, unverified scope) for a human reviewer — it never produces a pass/fail verdict.
    0
    installs
  3. ▌
    Desktop Test With Report · tnagatomi bundle
    Drive computer-use desktop automation (screenshot + click + type) to exercise the features a pull request changes in a native desktop app, then post an evidence-only findings report back to the PR. Sibling of agent-test-with-report — use THAT one for browser and Electron apps (they expose CDP, so agent-browser drives them); use THIS one for native, non-Chromium apps — Cocoa/SwiftUI/AppKit, Qt, Win32, and native-webview frameworks like Wails or Tauri on macOS (WKWebView), where agent-browser can't attach. Use when the user wants an agent to manually test a desktop-app PR on the real desktop and leave a report, with phrases like "agent-test this desktop app PR", "computer-use test the native app and comment on the PR", "run exploratory testing on the desktop build and post a report". macOS-local and interactive — not a CI gate. Collects evidence (operations, screenshots, findings, unverified scope); never a pass/fail verdict.
    0
    installs