
Sage Code
- Updated April 15, 2026
- umxr/sage-code
sage-code is a Claude Code skill in the AI & Agent Building category. Autonomous self-improving learning system for Claude Code — captures session events, reflects on outcomes, builds project-scoped knowledge, and self-evaluates its own rules
Key points
- sage-code
- AI & Agent Building
- AI-coding skill
Sage Code by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
/plugin marketplace add umxr/sage-code/plugin install sage-code@sage-code-marketplaceAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Last updated | April 15, 2026 |
|---|---|
| Repository | umxr/sage-code ↗ |
What it does
Autonomous self-improving learning system for Claude Code — captures session events, reflects on outcomes, builds project-scoped knowledge, and self-evaluates its own rules
README.md
sage-code
Self-Adapting Generative Engine for Code — a Claude Code plugin that makes Claude autonomously learn and improve within each project.
What it does
SAGE-Code observes your Claude Code sessions and builds project-specific knowledge over time:
- Captures corrections, tool outcomes, and patterns during sessions
- Reflects on what worked and what didn't, extracting reusable heuristics
- Replays relevant knowledge at the start of each session
- Self-evaluates whether its learned rules actually help, pruning ineffective ones
Everything is fully autonomous — no manual intervention needed.
Installation
Enable the plugin in your Claude Code settings:
{
"enabledPlugins": {
"sage-code@your-marketplace": true
}
}
On your first session in any project, SAGE will automatically initialize a .sage/ directory and begin learning.
Commands
/sage-status— View what SAGE has learned about your project/sage-reflect— Manually trigger reflection on current session
How it works
- Hooks passively capture session events (corrections, tool outcomes, successes)
- Reflector analyzes events and extracts generalized heuristics with confidence scores
- Knowledge files accumulate in
.sage/knowledge/as categorized markdown - Replay injects only relevant heuristics at session start based on git context
- Meta-evaluator periodically scores rules and prunes ineffective ones
Project data
SAGE creates a .sage/ directory in your project:
knowledge/— Learned heuristics (committed to git, shared with team)events/— Raw session logs (gitignored, personal)meta/— Evaluation scores and config (committed to git)
Configuration
Edit .sage/meta/config.json to tune thresholds. See the spec for all options.