skills.rest
mandubian
mandubianCOMMUNITY
Explore all skills by this author →
Website𝕏@mandubianGitHub

memory-curator.default

Aggregate cross-session traces to distill learnings and per-agent performance signals.

1|Updated Mar 23, 2026
Search Tags:#performance#evaluation#memory#cross-session#agent-evolution#pattern-learning
One-click install
npx skills add https://github.com/mandubian/autonoetic --skill memory-curator-default
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-curator.default
Source: https://github.com/mandubian/autonoetic/tree/main/agents/evolution/memory-curator.default
Command: npx skills add https://github.com/mandubian/autonoetic --skill memory-curator-default

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Aggregates cross-session traces to distill actionable learnings and per-agent performance signals, enabling focused evolution of capabilities.

Core Features & Use Cases

  • Cross-session learning distillation to extract reusable tactics and patterns.
  • Per-agent performance scoring across multiple signals to surface improvement opportunities.
  • Global knowledge storage of durable learnings for all agents to reference.

Quick Start

Run memory-curator.default on a batch of completed sessions to generate an evolution report.

Frequently Asked Questions about memory-curator.default

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

FAQPage Schema
How do I distill actionable learnings from multi-session agent traces?▼

Per-agent performance scoring evaluates execution traces using multi-signal metrics to compute scores, surface capability gaps, and generate targeted evolution recommendations with provenance tracking.

How does cross-session pattern learning work for agent evolution?▼

Cross-session pattern learning works by applying durable storage to learned patterns globally, computing multi-signal metrics from execution traces, and outputting evolution recommendations with provenance.

Can I use cross-session learning distillation for agent ecosystems with multiple sessions?▼

Run memory curation on a batch of completed sessions to generate an evolution report containing actionable learnings, per-agent performance signals, and recommended capability improvements.

What's the best way to generate an evolution report for agent capabilities?▼

Cross-session pattern discovery identifies reusable tactics across agent sessions by aggregating execution traces into globally stored durable learnings, enabling focused evolution of agent capabilities.

Do I need completed sessions to compute multi-signal performance metrics?▼

Yes, you need completed sessions because multi-signal performance metrics are computed from execution traces, which require finished agent sessions to yield actionable learnings and accurate scoring.

Related Skills

Explore More Software Engineering Skills→
unclehowellunclehowell

2nd-brain

Store durable facts and synchronize memory across agent sessions.

Community
Advanced
simotasimota
68

darwin

Detect project lifecycles and assess AI agent fitness for evolutionary actions.

Community
Advanced
alvarovillalbaaalvarovillalbaa

continuous-learning

Capture and synthesize engineering knowledge into project documentation files.

Community
Advanced
githubgithub
37.6k

agentic-eval

Automate iterative evaluation and refinement of AI agent outputs with JSON feedback.

Official
Intermediate