Item detail

regent-vcs/re_gent

regent-vcs/re_gent is a git-style version control for ai that RepoRadar is tracking in its Apache-2.0 `rgt` Go CLI that auto-captures every section, currently rated Gold tier with a 'try now' verdict. Its strongest signal is workflow potential, scored 9.2 out of 10.

Score8.1
Popularity754.0
Risklow
TierGold
Score breakdown
Usefulness8.0
Novelty9.0
Momentum7.0
Maturity8.9
Open-source/build8.4
Evidence7.2
Workflow potential9.2
Setup ease8.8

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for developers running AI coding agents (Claude Code, Codex, OpenCode) who need to answer, after the fact, 'which prompt wrote this line?', 'what did the agent do during this 2-hour session?', and 'why is this file like this?' — and who want a Git-style audit trail that is independent of the agent's own transcript log: a Go CLI that auto-captures every agent turn as a tracked step, `rgt log

Who should use it

Developers running AI coding agents (Claude Code, Codex, OpenCode) who need to answer, after the fact, 'which prompt wrote this line?', 'what did the agent do during this 2-hour session?', and 'why is this file like this?' — and who want a Git-style audit trail that is independent of the agent's own transcript logEngineering teams that need an audit trail for compliance / code review / incident response on AI-generated codeUsers debugging an agent regression who need to walk the chain of changes in reverse to find the prompt that introduced the bugMulti-agent workflows (concurrent Claude Code + Codex + OpenCode sessions on the same repo) where `rgt sessions` and `--session <filter>` are essential to disambiguate activityUsers who want the audit trail to outlive the agent's own transcript (the agent's session log is per-session and per-agent; regent persists across sessions and across agents)Anyone adopting the maintainer's `examples/bad-refactor/` worked debugging transcript as a reference for regent-driven debugging on their own codebase

Who should skip it

Skip regent-vcs/re_gent unless the captured evidence suggests it solves a problem you are actively working on.

About this signal

regent-vcs/re_gent is tracked by RepoRadar as a git-style version control for ai in the Apache-2.0 `rgt` Go CLI that auto-captures every section. It was first seen on 2026-06-25 and last updated on 2026-06-25. The current verdict is 'try now' with a Gold tier and easy setup difficulty. regent-vcs/re_gent leads on workflow potential (9.2) and novelty (9.0); its lowest signal is momentum (7.0), so factor that in before investing setup time. This page summarizes the evidence RepoRadar has captured from captured source metadata. The score, tier, risk label, and verdict on this page are never influenced by sponsorship, ads, or tips — they reflect only the usefulness, popularity, novelty, momentum, maturity, and evidence signals described in the RepoRadar methodology.

How this item is evaluated

RepoRadar assigned regent-vcs/re_gent a composite score of 8.1 out of 10, placing it in the Gold tier. This score combines weighted sub-signals: usefulness (35%), novelty (18%), momentum (14%), maturity (10%), open-source/build quality (7%), evidence quality (6%), workflow potential (6%), and setup ease (4%). Popularity is tracked separately at 754.0 and never affects the composite score or tier. The risk label of 'low' reflects inherent user-impacting hazards, not generic novelty. Items with no risk flag may still require normal code review before production use.

Risk explanation

**Captures the full prompt text that drove each change.** The `rgt blame` and `rgt show` outputs include the prompt text the user typed into the agent — users working on repos with sensitive context (internal system prompts, customer data pasted into prompts, pre-release code names) should review what the agent session log captures and decide whether the audit trail belongs in the repo, in a separate audit database, or in a private local store; the maintainer does not ship a built-in redaction layer, so users who need to redact should add it via the project's contributing path or store the trail outside the repo; **Captures multi-agent concurrent sessions under one trail but does not enforce cross-session conflict resolution.** `rgt sessions` lists concurrent Claude Code + Codex + OpenCode sessions, and `rgt log --session <id>` filters to one, but the step graph is shared across sessions — two agents editing the same file in the same window produce interleaved steps without a 'last writer wins' marker, so users running concurrent agents should pick one agent per working copy or use git worktrees to keep the steps cleanly separated; **Compatibility badges cover Claude Code, Codex, and OpenCode explicitly.** The README badges three agents (`Claude Code Compatible`, `Codex Compatible`, `OpenCode Compatible`); users on other agents (Cursor's in-editor agent, Windsurf, Devin, Aider, Continue) will need to verify the regent hook layer hooks into their agent's tool-call surface — if it does not, `rgt log` will be empty even though the agent is active.

Evidence links

Closest alternatives / related signals

regentre_gentregent-vcsrgtgo-clisingle-binaryhomebrew-tapclaude-code