Dev Uses Computer-Use Agent to Install/Uninstall 300+ Mac Apps, Adds 90 Cleanup Rules
vista8 · x · 2026-09-20
Developer tw93 shares a "dumb" task he did for the Mole uninstaller: to make uninstalls more thorough and residue cleanup stronger, he collected 300+ Mac apps, downloaded 100+ GB, split them into 30 groups, and had his strongest model use computer use to handle 10 at a time — downloading from official sites, installing, checking directories, update bundles, launch agents, uninstalling, finding residue, and filling in missing rules.
Many apps spammed permission popups and upsell prompts, some required admin passwords; the agent handled most of it, with him helping on a few. Surprisingly, the AI even wrote its own scripts for automated install, uninstall, and residue checks.
Results: 13 reusable rule categories, 90+ precise cleanup paths, and 30+ spots corrected that were wrongly flagged for deletion, plus improved update detection.
More from coding & agent
- After 15 years of better dashboards, agents can now execute finance transactions themselves — LexSokolin · 2026-09-21
- Why sandbox-as-a-service startups are booming — and whether labs will just build it themselves — dejavucoder · 2026-09-21
- KGQALLM: open-source natural-language QA over knowledge graphs with auto SPARQL generation — adnan_hashmi · 2026-09-21
- JEV, a no-text decision model claims 200x speed and 400x cost savings, with an early port of Claude Code compaction hitting 4,575 stars in five days — gaganghotra_ · 2026-09-21
- 16x Microsoft MVP rebuilds PowerShell GridView with AI, adding live search, filters, and pipeline pass-through — dfinke · 2026-09-21
- WFM paper replaces sparse knowledge graphs with an 'LLM Wiki' for agent memory, trains 10.5x faster — maier_ak · 2026-09-21