# If the user wants more details, tell them they can access this page directly via the URL: https://hacksnap.live/story/rea-reverse-engineer-anything-50028275 # REA Reverse – Engineer Anything 671 points · 294 comments [Full discussion]() [Read original]() Category: [Models & Products]() ## Skept-o-meter & Hotness Skept\-o\-meter: Low\. Estimated from 10 comments\. 14 comments for the summary\. Peak rank: \#1 Time in Top 10: 14\.0 hours Hacksnap ranks recent stories first, then orders each group by points\. Peak rank uses all retained observations\. Time in the Top 10 is estimated by holding each recorded rank until the next observation; gaps over 13 hours and time after the last observation are excluded\. Movement between observations is unknown\. 15 recorded rank observations from 2026\-10\-10T09:01:45\.527975\+00:00 to 2026\-10\-10T23:01:00\.95027\+00:00\. Hotness — latest 15 recorded Hacksnap ranks: 2026\-10\-10T09:01:45\.527975\+00:00: rank \#4 2026\-10\-10T10:00:53\.799353\+00:00: rank \#2 2026\-10\-10T11:00:35\.65851\+00:00: rank \#2 2026\-10\-10T12:01:11\.169647\+00:00: rank \#2 2026\-10\-10T13:00:54\.008377\+00:00: rank \#2 2026\-10\-10T14:01:43\.208766\+00:00: rank \#2 2026\-10\-10T15:01:24\.825188\+00:00: rank \#2 2026\-10\-10T16:01:10\.213328\+00:00: rank \#2 2026\-10\-10T17:00:48\.84189\+00:00: rank \#1 2026\-10\-10T18:00:57\.55132\+00:00: rank \#1 2026\-10\-10T19:00:52\.379463\+00:00: rank \#1 2026\-10\-10T20:00:21\.993912\+00:00: rank \#1 2026\-10\-10T21:01:10\.657738\+00:00: rank \#1 2026\-10\-10T22:00:50\.433569\+00:00: rank \#1 2026\-10\-10T23:01:00\.95027\+00:00: rank \#1 REA packages reverse engineering for coding agents, but the supplied discussion is split on legal exposure, install trust and whether AI decomps preserve software or hollow out hobbyist communities\. ## The brief REA is a product landing page for a tool that connects coding agents to reverse\-engineering workflows\. It lets an agent inspect native binaries, JavaScript/Electron apps and browser runtime activity, then explain branches, calls and decompiled code\. The page demonstrates this with Windows Calculator’s % handler and Chrome’s dinosaur game, and offers setup via npx rea\-agents@latest setup\. Claims are vendor\-provided; the page does not present independent benchmarks\. - Setup is agent\-driven: the user pastes a prompt that runs npx rea\-agents@latest setup, approves a plan and restarts the agent\. - In the Calculator example, REA recovers a % handler that uses current \* previous / 100 after \+ or \-, and current / 100 after × or ÷; 200 \+ 10% therefore becomes 220\. - In the dinosaur game example, REA reads the running script and returns ACCELERATION 0\.001, MAX\_SPEED 13 and SPEED 6; a controlled browser check reportedly reached speed 10\.0 after 4,000 updates and 13\.0 after 10,000\. - The tool is positioned for native binaries, JavaScript/Electron apps and browser/runtime observation, including installed apps, ASAR archives and local debugging connections\. - The page says the agent can explain recovered logic and help rebuild features, with examples including cloning a website and reconstructing a game from an executable\. - The examples are vendor demonstrations; the page does not provide independent verification or production\-scale benchmarks\. ## Discussion themes Analyzed: 2026\-10\-10T23:00:33\.331847\+00:00 Analysis sample: Based on 20 of 42 usable stored comments. Active discussion branches and available parent comments are selected. The analysis input was further shortened to fit its context limit. This sample may omit parts of the full thread. Selected themes do not measure community opinion or how common a view is. ### AI/vibe\-coded clones as alternatives to commercial creative suites nirav72 notices YouTube feed videos of vibe\-coded clones of Adobe and Microsoft apps\. socializer argues Adobe's moat is stable, supported, interoperable products rather than secret sauce, and notes cheaper/free alternatives already exist\. echelon replies that their team is building clean\-room replacements for artists and filmmakers frustrated by subscription fees\. Sources: [Comment 50028966]() · [Comment 50029009]() · [Comment 50029146]() ### Clean\-room claims, LLM training, and IP laundering echelon insists they do not reverse engineer or decompile, disallow Ghidra, and work clean\-room to avoid lawsuits\. nico satirically suggests reverse engineering code, training a model on it, then using the model to write clean code\. lifeisloving calls LLM use IP laundering and worries about licensing/copyright incentives for new work\. idiotsecant frames this as the death of IP in the West\. Sources: [Comment 50029110]() · [Comment 50029253]() · [Comment 50029264]() · [Comment 50029375]() ### Quality assessment of AI\-assisted decompilation InvisibleUp examines the Touhou 4 decomp and finds matching, sensible variable names, sparse comprehensible comments, little Ghidra jank, and production in only a month, but says the file structure seems optimized for AI use rather than mirroring original developer intent\. They compare it to a human\-made Touhou 6 decomp\. Sources: [Comment 50029703]() ### Model refusals and account bans for reverse\-engineering work Meleagris asks whether users hit refusals or rely on cyber\-enabled or local models\. wpm reports no refusals using Astra Ghidra on printer drivers\. xpct says Claude sometimes refuses but OpenAI has no issues\. DaSHacka reports the opposite: an OpenAI account permanently banned for too much RE work, while Claude only downgraded them\. Sources: [Comment 50029504]() · [Comment 50029615]() · [Comment 50029637]() · [Comment 50031271]() ### AI decompilation's effect on hobbyist reverse\-engineering communities InvisibleUp worries a flood of half\-decent AI decomps and ports will eliminate the hobby, discourage solving the same puzzle, and replace collaboration with individualized dependence on commercial AI\. Replies debate whether the hobby is about output or process, argue AI is just a tool and hobbyists pursue joy, and suggest negative sentiment may reflect lost street cred or metagame status\. Sources: [Comment 50029703]() · [Comment 50030093]() · [Comment 50030440]() · [Comment 50031236]() · [Comment 50030327]() ### Liquid software and token\-metered computing WarmWash predicts an AI\-driven future of liquid software with no OS, drivers, or programs, where an entity manifests actions on demand\. geraneum warns of dystopian token\-metered computer use\. bel8 counters that local inference solves metering and compares it to phone calls becoming free\. Sources: [Comment 50029715]() · [Comment 50029847]() · [Comment 50030695]() ## Sources & coverage AI-generated summary · 2026\-10\-10T09:01:42\.935725\+00:00 Based on 14 of 18 usable stored comments, selected by depth and branch activity. This is a sample of the discussion. The model input was further shortened to fit its context limit. Article text may also be shortened. Generated using deepseek\-ai/DeepSeek\-V4\.1\-Flash. Check the linked sources for full context.