realisera

Automate iterative codebase evolution through structured development cycles with guardrails.

3|Updated Mar 15, 2026
One-click install
npx skills add https://github.com/jgabor/agentera --skill realisera
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: realisera
Source: https://github.com/jgabor/agentera/tree/main/skills/realisera
Command: npx skills add https://github.com/jgabor/agentera --skill realisera

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automates continuous, autonomous development cycles for a codebase, enabling a single agent to decide what to build, evolve a project over time, and progress through structured development cycles with safety rails and decision-driven governance.

Core Features & Use Cases

  • Autonomous decision-making to select and execute development cycles
  • Structured loops with vision-driven guidance and safety rails
  • Continuity of project state across cycles via documented artifacts

Quick Start

Trigger a cycle by saying realisera or starting a loop to begin an autonomous development cycle.

Frequently Asked Questions about realisera

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate autonomous development cycles for a codebase?

Autonomous development cycles automate iterative codebase evolution by guiding a system to independently decide what to build and execute structured loops. The system reads VISION.md and TODO.md to select tasks, then updates PROGRESS.md and CHANGELOG.md as it progresses.

What is autonomous looped software development and how does it work?

Looped software development continuously evolves a project through iterative cycles. The system reads identity from SKILL.md frontmatter, selects tasks from TODO.md, executes them with safety guardrails, and records state changes in PROGRESS.md and CHANGELOG.md for continuity.

Can I use persona-driven decisions and safety guardrails for continuous codebase evolution?

Yes, continuous codebase evolution supports persona-driven decisions and safety guardrails. The system applies decision profiles and risk-safety constraints during each cycle, ensuring autonomous development progresses within defined boundaries documented in HEALTH.md.

Do I need specific markdown files to start autonomous project evolution?

Yes, autonomous project evolution requires VISION.md, PROGRESS.md, TODO.md, and HEALTH.md to function. The system reads these files to understand project identity and state, then updates them alongside CHANGELOG.md as development cycles progress.

What's the best way to maintain project state continuity across autonomous development cycles?

Project state continuity is maintained by storing ongoing development state in VISION.md, TODO.md, PROGRESS.md, and CHANGELOG.md. The system reads and updates these artifacts each cycle, ensuring progress persists across autonomous loops without losing context.

When should I not use autonomous development loops for software evolution?

Autonomous development loops are not suitable when a project requires manual oversight for each change or lacks initial documentation in VISION.md and TODO.md. Without established safety guardrails and decision profiles, the system cannot reliably select and execute development cycles.