Coding-agent monitoring

    Stop re-explaining the same thing to your agents

    Hervé reads your whole team's agent sessions to find and fix what keeps going wrong in your software factory.

    Run a free scan

    Free. Runs on your machine. No account.

    FindFixGuardHervé spots where your agents keep making the same mistake.
    acme/platform · last 30 days

    5 findings across 412 sessions

    14 developers · Claude Code, Codex

    More agents. More rules. Less control.

    Your CLAUDE.md keeps growing, and nobody knows which rules the agents actually follow.

    The same instructions get retyped across the team, when they should be one shared skill.

    A secret pasted into a prompt, or printed by an agent, is gone without anyone even noticing.

    No one sees it from inside a single session. Hervé reads all of them.

    Your first six weeks with Hervé

    1. Day 1 · Install

      Each team member runs rv install once. Their agent sessions are captured to a branch in your own repo.

    2. Week 2 · Findings

      We walk you through what Hervé found across your team's sessions, and open the first fixes.

    3. Weeks 3–6 · Guard

      Hervé checks every new session against your rules, with a weekly update.

    See pilot details & pricing →

    Questions

    Where does our code go?

    Nowhere. Hervé reads code on demand and never stores it: no diffs, no file contents in our database.

    Where do our AI sessions go?

    Into your own repo. Sessions are saved to a branch of your repository, only for the repos where your team ran rv install. We only store aggregate findings and counters. Security details →

    What does each developer have to install?

    One command, run once: rv install. Every team member runs it, because Hervé only sees the sessions it captures.

    Do you integrate with our stack?

    Hervé works with GitHub and GitLab, reads the Jira tickets your PRs mention, and captures sessions from Claude Code, Cursor, GitHub Copilot CLI, OpenAI Codex, and OpenCode. Setup is one install script and one command per repo.

    Can't I just ask Claude to clean up my CLAUDE.md?

    You can, for one session. Claude sees what happened in front of it; it can't see that three developers hit the same wall last week, or which rules no agent has read in a month. Hervé reads every session across the team, then checks whether the fix held.

    What happens after the six weeks?

    You continue with a monthly license, or stop and we delete your data. See pilot details & pricing →

    See what your team's sessions are hiding.

    Run a free scan

    Free. Runs on your machine. No account.