debug-agent

Name: debug-agent
Rating: 5 (2 reviews)
Author: subarnasaikia

Step through a simulation agent replay to find where and why it failed — timeout, invalid move, crash, or strategic collapse. Use when the user says "debug the agent", "why did it lose", "find the failing turn", "analyze the replay", or after a regression in local eval.

2stars

Updated 2 months ago

View on GitHub ↗License: MIT

How to add

/plugin marketplace add subarnasaikia/kaggle-skills

The exact command may vary by repository. Check the README on GitHub.

For the skill author

Drop this on your repo README

Shows your skill is listed on Skillteca, generates a backlink and trackable traffic.

[![Listada na Skillteca](https://www.skillteca.com.br/api/badge/debug-agent/svg)](https://www.skillteca.com.br/skills/debug-agent?utm_source=badge&utm_medium=readme&utm_campaign=badge)

#ai

Related skills

See all in Desenvolvimento →

claude-api

143.8k

Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. Also handles migrating existing Claude API code between Claude model versions (4.5 → 4.6, 4.6 → 4.7, retired-model replacements). TRIGGER when: code imports `anthropic`/`@anthropic-ai/sdk`; user asks for the Claude API, Anthropic SDK, or Managed Agents; user adds/modifies/tunes a C

Desenvolvimento#ai#apiby anthropics

skill-creator

143.8k

Create new skills, modify and improve existing ones, and measure their performance. Use this skill for developing, editing, optimizing, testing, and benchmarking skills, as well as refining their descriptions for better triggering.

Desenvolvimento#testby anthropics

oh-my-issues

79.7k

This skill clusters GitHub issue backlogs by root cause into plan-master issues, redirects related children, and bundles architectural-fix PRs to close clusters atomically. It's ideal for triaging and consolidating numerous issues sharing underlying defects, or for building a plan series or roadmap.

Desenvolvimento#github#gitby thedotmack

claude-mem

79.7k

Captures agent actions across sessions, compresses them with AI, and injects relevant context into future interactions. Compatible with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode, and more.

Desenvolvimento#aiby thedotmack

Category alert

Get new Desenvolvimento skills every Monday

One short email with only the new Desenvolvimento skills. 4 minutes of reading, no spam, unsubscribe with one click.

You confirm your email on the first send. No spam. Unsubscribe with one click.

Debug Agent — Replay Forensics

Turns a raw replay JSON into a structured diagnosis: which turn failed, what state the agent saw, what it returned, and why.

When to use

Local eval showed a loss, crash, or invalid move.
A submission scored worse than expected; you have a replay to examine.
An agent works in 1v1 but fails in FFA (or vice versa).
A specific seed consistently loses; others are fine.

Steps

1. Locate the replay

ls competitions/<slug>/replays/
# pick th

[Description truncada. Veja o README completo no GitHub.]

ShareX LinkedIn

Comments · No comments

No comments yet. Be the first.