Veronica — AI skill for Claude Code
Veronica, an optimized Ruflo.
How to install Veronica
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open joaoreislab/veronica and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Veronica does
Veronica, an optimized Ruflo? An agent skill for long-running AI work with local state, dependencies, resumption, and completion evidence.
Alternatives in AI
- Ralph For Claude Code — by Frank Bria - An autonomous AI development framework that enables Claude Code to work iteratively on project 8k ★
- Hindsight — State-of-the-art long-term memory for AI agents by Vectorize 6.7k ★
- Claude Squad — by smtg-ai - Claude Squad is a terminal app that manages multiple Claude Code, Codex (and other local agents i 6.5k ★
README

Veronica, an optimized Ruflo?
**🌸 Give a long AI project a clear path forward, a place to save its progress, and a way to check the result.**
Ever returned to an AI project and spent the first hour explaining it all over again? Veronica helps your assistant keep the goal, detailed requirements, decisions, unfinished work and next steps in your workspace, ready for the next session.
**Veronica 1.2.0** is a reusable skill with lightweight local tools. Your assistant does the thinking and uses the tools available in its environment; Veronica gives that work continuity and structure.
[🚀 Get started](#-get-started) · [🌸 What's new](#-whats-new-in-120) · [📊 Measured results](#-what-we-actually-measured) · [📣 Announcement](docs/ANNOUNCEMENT.md)
🧭 Built for work that takes more than one sitting
Research a difficult topic. Build an app. Prepare a report with dozens of sources. Improve a project over several rounds of feedback.
Veronica helps turn that large objective into useful tasks, tracks which tasks depend on others, and records discoveries as the work evolves. It saves a handoff with what has been done, what remains uncertain and what to do next. When you return, the assistant can check the actual files and continue from that record.
Detailed inputs and original files remain available alongside shorter views. A handoff is a route back to the work, not a replacement for its full history. The model still has a finite context window, so continuity depends on preserving and consulting those records.
🤝 A team when the project benefits from one
When your environment supports agents and their use is authorized, Veronica guides the assistant in dividing independent work: one agent researches, another implements, another reviews.
Each assignment gets an expected deliverable, completion criteria and a declared area for writing. Cooperative reservations help avoi
Related Skills
Canny
Stops AI coding agents from claiming work is done without evidence. Deterministic hooks decide, TypeSafe's Jev
Cwc Long Running Agents
Claude Code's built-in /goal command gives you a generator/evaluator loop out of the box: set a completion con
Ralph Loop
A long-running AI agent loop. Ralph automates software development tasks by iteratively working through a task
AI Dev Orchestrator Complete
A local-first, multi-agent software development operating layer that routes work across Codex, Claude Code, Cu
Late CLI
High-performance AI agent for long-horizon tasks and local models. Built on empirical research. 200k+ tokens o
Ante
Ghost in your shell. Ante is a self-contained agent harness with a highly optimized core. It works like Claude
Related Agents
Project Tracker
Save current Claude Code session as a project in Second Brain for later resumption. Gathers git state, files t
Hatch3r Handoff Loader
Session-start agent that surfaces active handoff documents from .hatch3r/handoffs/active/. Use at the beginnin
Session Log.Agent
Use at the end of a work session to append a structured session log to ~/.local/state/session-logs/YYYY-MM-DD.