Version-Control for AI coding agents.
$ npx -y skills add regent-vcs/re_gent --agent claude-code
Run the curl in your terminal, the rest in Claude Code.
What's inside
# Install via Homebrew (macOS/Linux)
brew tap regent-vcs/tap
brew install regent
# Or via Go
go install github.com/regent-vcs/regent/cmd/rgt@latest
# Initialize in your project
cd your-project
rgt init
# Work with Claude Code, Codex, or OpenCode normally โ activity is tracked automatically
# See what happened
rgt log
rgt blame src/file.go:42
rgt show <step-hash>
That's it. Your agent activity is now auditable.
https://github.com/user-attachments/assets/a19b7c56-2e3c-4f04-81a1-d8665e3963b8
rgt log, rgt blame, and rgt show.$ rgt log
Step a1b2c3d | 2 min ago | Tool: Edit
โ File: src/handler.go
โ Added error handling to request handler
โ + 5 lines, - 2 lines
Step d4e5f6g | 5 min ago | Tool: Write
โ File: tests/handler_test.go
โ Created unit tests for handler
โ + 23 lines
Step f8g9h0i | 8 min ago | Tool: Bash
โ Command: go mod tidy
โ Cleaned up dependencies
$ rgt blame src/handler.go:42
Line 42: func handleRequest(w http.ResponseWriter, r *http.Request) {
Step: a1b2c3d4e5f6
Session: claude-20260502-143021
Tool: Edit
Prompt: "Add error handling to the request handler"
$ rgt sessions
Active Sessions:
claude_code:claude-20260502-143021 | 3 steps | Last: 2 min ago
codex_cli:codex-20260502-091534 | 7 steps | Last: 2 hours ago
$ rgt log --session claude_code:claude-20260502-143021
# Filter history by session
$ rgt show a1b2c3d
Step a1b2c3d4e5f6
Parent: d4e5f6g7h8i9
Session: claude-20260502-143021
Time: 2026-05-02 14:30:21
Tool: Edit
File: src/handler.go
Changes:
+ func handleRequest(w http.ResponseWriter, r *http.Request) {
+ if r.Method != "GET" {
+ http.Error(w, "Method not allowed", 405)
+ return
+ }
- func handleRequest(w http.ResponseWriter, r *http.Request) {
Conversation:
User: "Add error handling to reject non-GET requests"
Assistant: "I'll add method validation to the handler..."
AI agents have no version control of their own.
You know this pain:
/compact and prayThree primitives that should already exist:
rgt log โ what did this session do?rgt blame โ which prompt wrote this line?rgt show โ inspect the full context for any stepWe gave agents write access to our codebases. We did not give ourselves git for it. re_gent fixes that.
re_gent stores agent activity in .regent/ (like .git/):
.regent/
โโโ objects/ # Content-addressed blobs (BLAKE3)
โโโ refs/ # Session pointers (one per agent)
โโโ index.db # SQLite query index
โโโ config.toml
Every tool-using turn creates a Step โ a content-addressed snapshot of what changed, why, and who asked:
Step {
parent: <previous-step-hash>
tree: <workspace-snapshot>
causes: [{ tool_name: "Edit", args: <input>, result: <output> }]
session_id: "claude_code:claude-20260502-143021"
timestamp: "2026-05-02T14:30:21Z"
}
Steps form a DAG. Each session has its own branch. Common ancestors dedupe. You get git-level auditability for agent activity.
Technical details: See POC.md for the complete specification.
brew tap regent-vcs/tap
brew install regent
This installs the rgt command and automatically sets up shell completions for bash, zsh, and fish.
go install github.com/regent-vcs/regent/cmd/rgt@latest
Shell Completion (manual setup):
# Bash
rgt completion bash > /usr/local/etc/bash_completion.d/rgt
# Zsh
rgt completion zsh > "${fpath[1]}/_rgt"
# Fish
rgt completion fish > ~/.config/fish/completions/rgt.fish
git clone https://github.com/regent-vcs/regent
cd regent
go build -o rgt ./cmd/rgt
sudo mv rgt /usr/local/bin/
Download pre-built binaries from GitHub Releases
| Tool | Status |
|---|---|
| Claude Code | Fully supported |
| OpenAI Codex CLI | Fully supported |
| OpenCode | Fully supported |
| Cursor, Cline, Continue | Planned |
Hooks auto-configure on rgt init. No manual setup required.
| Command | Description |
|---|---|
rgt init | Initialize .regent/ in current directory |
rgt log | Show step history (supports --session, -n, --json, --graph) |
rgt sessions | List all active sessions |
rgt status | Show current repository state |
rgt show <step> | Display full context for a step (tool call + conversation) |
rgt blame <path>[:<line>] | Show per-line provenance for a file |
rgt cat <hash> | Inspect any object by hash |
rgt version | Print version information |
rgt completion | Generate shell completion scripts |
/compact and /clearrgt init.regentignore supportGet inline blame annotations directly in your editor:
# From VSIX (Recommended)
# Download the latest .vsix from:
# https://github.com/regent-vcs/vscode-regent/releases
# Then in VS Code: Extensions > ... > Install from VSIX...
# From source (Development)
git clone https://github.com/regent-vcs/vscode-regent
cd vscode-regent
npm install && npm run compile
# Press F5 in VS Code to launch Extension Development Host
Features:
Requirements: rgt CLI must be installed and rgt init run in your project.
| Git | re_gent | |
|---|---|---|
| Tracks code | โ | โ |
| Tracks agent activity | โ | โ |
| Blame with prompt | โ | โ |
| Conversation history | โ | โ |
| Concurrent sessions | โ ๏ธ shared workspace conflicts | โ separate captured session refs |
| Purpose | Developer VCS | Agent audit trail |
re_gent complements git, doesn't replace it. Use both.
See ROADMAP.md for planned features including:
Contributions welcome! See CONTRIBUTING.md for guidelines.
Quick Start:
Before opening a PR:
go test ./... and go test -race ./...golangci-lint rungo fmt ./....agents/
skills/
blame/
SKILL.md
log/
SKILL.md
show/
SKILL.md
.claude/
settings.json
skills/
blame/
SKILL.md
log/
SKILL.md
rewind/
SKILL.md
show/
SKILL.md
.coderabbit.yaml
.github/
dependabot.yml
GOOD_FIRST_ISSUES.md
ISSUE_TEMPLATE/
bug_report.yml
feature_request.yml
phase_implementation.yml
MORE_ISSUES.md
pull_request_template.md
QUICK_START.md
workflows/
ci.yml
release.yml
.gitignore
.golangci.yml
.goreleaser.yaml
AGENTS.md
assets/
demo-fast.gif
demo-trimmed.gif
demo.gif
demo.mp4
regent-icon.svg
regent-logo-dark.png
BRAND.md
CHANGELOG.md
CLAUDE.md
cmd/
rgt/
codex_hook_test.go
codex_hook.go
main.go
message_hook.go
opencode_hook_test.go
opencode_hook.go
pi_hook_test.go
pi_hook.go
tool_batch_hook.go
CODE_OF_CONDUCT.md
CONTRIBUTING.md
CONTRIBUTORS.md
demo/
.claude/
settings.json
skills/
blame/
SKILL.md
log/
SKILL.md
show/
SKILL.md
greet.py
hello.py
opencode.jsonc
docs/
FAQ.md
images/
log-command-example.png
shell-completion.md
examples/
bad-refactor/
.gitignore
app_before.py
app.py
Dockerfile
README.md
regression_test.py
setup.sh
go.mod
go.sum
internal/
capture/
capture_test.go
capture.go
cli/
blame_test.go
blame.go
cat.go
graph_renderer_test.go
graph_renderer.go
hook.go
init_test.go
init.go
log_enrich_test.go
log_enrich.go
log_format_test.go
log_format.go
log.go
message_format.go
repo.go
sessions_test.go
sessions.go
show_test.go
show.go
status_test.go
status.go
version_test.go
version.go
conversation/
converter_test.go
converter.go
diff/
diff_test.go
diff.go
hook/
hook_test.go
hook.go
payload.go
ignore/
ignore_test.go
ignore.go
index/
index_test.go
index.go
messages_test.go
messages.go
resolve.go
jsonl/
extract_test.go
extract.go
snapshot/
snapshot_test.go
snapshot.go
store/
blame_test.go
blame.go
blob_test.go
blob.go
hash_test.go
hash.go
refs_test.go
refs.go
step_test.go
step.go
store.go
transcript_test.go
transcript.go
tree_test.go
tree.go
workspace_test.go
workspace.go
style/
style.go
treediff/
treediff_test.go
treediff.go
LICENSE
Makefile
POC.md
README.md
ROADMAP.md
scripts/
release.sh
test-release.sh
SUPPORT.md
test/
integration_test.go
phase1_acceptance_test.go
session_branching_test.go
TESTING.mdFAQ
re-gent is a Claude Code plugin with 7 hand-picked skills for development work, indexed on Flowy. Install it with the command on its page. It includes blame, log, rewind. Its skills do not fire on their own yet. Request auto-invocation to have Flowy route them as you prompt. Free and open source.