skills.rest
KingInYellows
KingInYellowsOFFICIAL
Explore all skills by this author →
GitHub

agent-learning

Record and retrieve agent learnings with RRF-based ranking and quality gates.

Updated Jan 10, 2026
Search Tags:#retrieval#memory#quality-gates#learn#agent-learning#reciprocal-rank-fusion#learning-entries
One-click install
npx skills add https://github.com/KingInYellows/yellow-plugins --skill agent-learning
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-learning
Source: https://github.com/KingInYellows/yellow-plugins/tree/main/plugins/yellow-ruvector/skills/agent-learning
Command: npx skills add https://github.com/KingInYellows/yellow-plugins --skill agent-learning

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Defines when to record learnings, quality standards for entries, and retrieval strategies. Loaded by memory-related commands and agents for consistent learning behavior.

Core Features & Use Cases

  • Learning triggers across contexts (context, decision, code entries) with structured guidelines.
  • Retrieval strategy using Reciprocal Rank Fusion (RRF) to rank learnings and manage a session budget.
  • Quality gates ensure entries meet minimum content length and actionability for future agents.

Quick Start

Record a context entry after a failure with clear context, insight, and action.

Frequently Asked Questions about agent-learning

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

FAQPage Schema
How do I standardize agent learnings and enforce quality gates in memory workflows?▼

You can standardize agent learnings by applying quality gates that enforce minimum content length and require explicit context, insight, and action for structured retrieval and memory workflows.

What is Reciprocal Rank Fusion used for when retrieving agent learnings?▼

Reciprocal Rank Fusion (RRF) is used to rank retrieved agent learnings and manage session budgets, ensuring memory workflows retrieve the most relevant and actionable context.

When should I record context, decision, and code entries for agent learning?▼

You should record context, decision, and code entries after meaningful events like failures or successful resolutions to capture clear context, insight, and action for future agent workflows.

How do I ensure recorded agent learnings remain actionable for future agents?▼

To ensure learnings remain actionable, apply quality gates that validate minimum content length and verify that each entry contains explicit context, insight, and action components.

Does this agent-learning approach work without external dependencies?▼

Yes, this approach operates without external dependencies, using built-in triggers and quality gates to standardize learning entries and manage retrieval workflows natively.

What are the limitations of managing agent learnings with session budgets?▼

Session budgets limit the number of learnings retrieved during a workflow, which requires precise RRF-based ranking to ensure only the most relevant entries consume the available context window.

Related Skills

Explore More Data & Analytics Skills→
ilandahanilandahan
9

phase-enforcement

Enforce phase gates and validate transitions across AI.D project stages.

Community
Advanced
ionmidoriionmidori

implementing-hybrid-search

Combine vector and keyword search results using Reciprocal Rank Fusion.

Community
Intermediate
Tx1207Tx1207
5

memory-updater

Synchronize the global memory file with repository changes in skills, commands, agents, and hooks.

Community
Intermediate
soo-kate-yeonsoo-kate-yeon

moai-foundation-context

Manages token budgets, session persistence, and multi-agent context transfers for workflows with strict constraints.

Community
Advanced