Open-source agent that takes a ticket to a reviewed pull request, with proof
Emerged finds new products worth a look, daily — what’s new today
Sentrint
Verified founderStatic security scan of a GitHub repo, with fix prompts for your AI coding tool
If your app was largely written by an AI tool, this gives a ranked list of what an attacker would try first and a prompt to hand back to that tool.
Sentrint reads a GitHub repository without running the app and looks for hardcoded secrets, weak database access rules, vulnerable dependencies and risky code paths. An AI pass rechecks a capped number of findings to cut false positives. Results come as a score and grade, findings in plain English with file and line, an attack path for each, and a fix prompt written for tools such as Cursor, Lovable or Claude Code. There is also an MCP server and REST API. One free scan a month; six more for $10, or $22 a month. Run by Gourab Dasgupta in Kolkata.
More like this
Mark up your coding agent's plans and diffs in the browser and send notes back
A benchmark where agents get only a repo and must find the bugs themselves
Watches production, finds why it broke and opens a pull request with the fix
Replays a pull request as the ordered steps that could have built it
A local firewall that stops AI agents acting on hostile content they just read