skills.rest
nahisaho
nahisahoCOMMUNITY
Explore all skills by this author →
GitHub

agent-assistant

Detect agent stuck patterns and loops and manage memory adjustments.

69|5|Updated Nov 16, 2025
Search Tags:#learning#memory#session#stuck-detection
One-click install
npx skills add https://github.com/nahisaho/musubi --skill agent-assistant
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-assistant
Source: https://github.com/nahisaho/musubi/tree/main/.claude/skills/agent-assistant
Command: npx skills add https://github.com/nahisaho/musubi --skill agent-assistant

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill acts as an assistant to diagnose agent stuck conditions, manage memory, and extract learnings for ongoing sessions.

Core Features & Use Cases

  • Stuck-detection (repeats, loops, context overflow)
  • Memory condenser and extractor
  • Session learnings persistence and retrieval

Quick Start

Run stuck-detection on the current agent session and cache learnings for reuse.

Frequently Asked Questions about agent-assistant

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

FAQPage Schema
How do I detect when an agent is stuck in a loop or repeating the same actions?▼

Stuck-detection identifies repeating patterns, loops, and context overflow in agent sessions. Run the detector on your current agent session to flag stuck conditions and receive remediation recommendations based on the detected pattern type.

What should I do when an agent reports being stuck or needs memory management?▼

Apply stuck-detection to diagnose the condition, then use the memory condenser to reduce session size and extract key learnings. The Skill determines whether to reset context, condense memory, or adjust session parameters based on the stuck pattern identified.

How can I extract and reuse learnings from agent sessions?▼

The Skill persists session learnings for retrieval and reuse across sessions. After running detection and memory management, cache extracted insights to improve agent behavior in future interactions without rebuilding context from scratch.

Can I integrate stuck-detection into my agent's CLI or API workflow?▼

Yes. The Skill provides CLI and API integration for stuck-detection, memory condensation, and learning extraction, allowing you to automate diagnostics and memory management within existing agent orchestration pipelines.

What causes agent context overflow and how does memory condensing help?▼

Context overflow occurs when session history exceeds available tokens or reasoning capacity. Memory condensing strategies reduce session size by summarizing non-critical events while retaining decision-critical information, freeing capacity for ongoing reasoning.

Related Skills

Explore More Software Engineering Skills→
matteingmatteing
56

debugging

Diagnose Opal agent runtime failures with structured debug_state JSON snapshots.

Community
Intermediate
mupengi-botmupengi-bot
11

mupeng-learn

Collect subagent learning reports and integrate insights into memory and knowledge bases.

Community
Advanced
kivo360kivo360
1

moai-alfred-session-state

Manage Alfred session state across workflows with session_state.json persistence.

Community
Advanced
Aurora-AIAurora-AI
1

Hive Mind Protocol

Coordinate distributed agents for parallel task execution and output aggregation.

Community
Advanced