Gabson0x/bountyforge ? reverse-engineered prompt

Build me a Claude Code skill for bug bounty work that can take a target like a web app, API, smart contract repo, or security scope and run a bunch of focused checks in parallel, then turn the results into a clean report I can submit.

I want it to look for the usual high value issues in web and API security, access control, business logic, crypto and math mistakes, race conditions, economic attacks, and recon findings. It should also handle smart contract audits for EVM, Move, Solana, and TRON, and support local tools when code execution is enabled, so it can use things like scanners and fuzzers if they are installed.

Please make the findings get deduplicated, scored, and filtered so only the real stuff makes it into the final writeup. I also want report output that can be formatted for HackerOne, Bugcrowd, Intigriti, or Immunefi, with an option to save the report to a file. If you need to check current docs or tool behavior online, go ahead and look them up.