Hacker

Sanath R

@sansynx

into the unknown

Joined Aug 2026unemployed at homeIndia

Elsewhere

Projects

  • Orbit: Keep track of every AI-assisted contribution

    No Hands: The Wispr ChallengeOct 2026

    When you contribute to open source, the work doesn’t end after raising a pull request. You still need to follow up on reviews, respond to maintainers, and remember what needs changing. Once you have several PRs across different repositories, keeping track of everything becomes difficult especially when you’re switching between coding agents like Codex, Claude Code, and Cursor. The context often stays inside separate agent conversations, so picking up an older contribution means digging through chats and GitHub tabs I built Orbit to make that easier. It’s a contribution tracker for developers who use AI coding agents. The Orbit CLI registers a PR along with the agent that worked on it, a short summary, tests, decisions, and the next action. A central dashboard brings together PRs, raised issues, review updates, and contribution insights. When you switch agents, the next agent can retrieve the saved context instead of starting from scratch. The goal is to spend less time remembering where you left off and more time contributing Wispr Flow was how I communicated throughout the build. I spoke my instructions instead of typing them from explaining the initial problem and planning the architecture to describing interface changes, reporting bugs, and asking the coding agents to test the application. Being able to talk through what I wanted made the process feel much more natural, particularly when a change was easier to explain aloud than write as a detailed prompt

  • Top 50

    Arcloom: Evidence That Survives Agent Handoffs

    Graph HacksOct 2026

    Security investigations often lose context when a session ends or another investigator takes over. A useful finding gets buried in a report, and its connection to the next investigation disappears. Arcloom turns reviewed security evidence into persistent, connected memory. A researcher submits a report, a human approves a bounded test, and a separate review determines whether its evidence becomes a confirmed finding. FalkorDB connects those findings to the capabilities they enable and the tests that require them. A fresh investigator can then propose a related test with an inspectable supporting path, without inheriting the previous conversation. Our demonstration follows a gateway finding into a separate invoice investigation and then a chain requiring both findings. It also preserves negative results, researcher attribution and deployment history. Every new execution requires approval, and a deployment change prevents outdated findings from silently authorizing new work. Built for security researchers, reviewers and teams developing accountable agent workflows.

    Into the unknown@mohiths