본문으로 건너뛰기
cccskills
무료GitHub에 공개

agent-repair

A two-phase repair skill that analyzes errors and suggests fixes before executing repairs. Phase one: user describes error, agent analyzes and proposes solution. Phase two: upon approval, executes the repair action.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

설치 방법 보기

포함된 파일(2)

  • SKILL.md5.5 KB
  • metadata.json1.4 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Agent Network Repair

Overview

[TODO: 1-2 sentences explaining what this skill enables]

Agent Network Repair Skill

Overview

This skill provides a two-phase approach to diagnosing and repairing system errors, service failures, and configuration issues.

Workflow

Phase 1: Analysis and Discussion

When the user reports an error or issue:

  1. Gather Information

    • Ask the user to describe the error, symptoms, and context
    • Collect error messages, stack traces, or failure logs
    • Identify which service, API, or component is affected
  2. Diagnose the Problem

    • Analyze the error type (authentication, configuration, network, service availability, etc.)
    • Check relevant logs or status if accessible
    • Identify potential root causes
    • Consider multiple possible solutions
  3. Present Initial Analysis

    • Clearly explain what the problem appears to be
    • List the likely root cause
    • Propose one or more repair actions
    • Explain potential risks or side effects
    • Invite questions or discussion
  4. Iterative Discussion Loop

    • User can ask follow-up questions about the analysis
    • User can request alternative solutions or approaches
    • User can ask for more details about specific aspects
    • User can request additional investigation or checks
    • Agent refines analysis based on feedback
    • Continue this back-and-forth until user is satisfied
    • Only proceed to Phase 2 when user explicitly approves a specific action

Discussion continues until user says: "proceed", "do it", "approved", "go ahead", or similar confirmation

Phase 2: Repair (Only After Explicit Approval)

Once the user approves the proposed solution:

  1. Execute Repair Actions

    • Implement the approved fix (e.g., update configuration, restart service, fix credentials)
    • Use appropriate tools (file edits, bash commands, API calls)
    • Log all actions taken
  2. Verify the Fix

    • Test that the issue is resolved
    • Confirm the service or component is working
    • Report results back to the user
  3. Document

    • Summarize what was done
    • Note any follow-up actions needed

Common Repair Scenarios

  • API Authentication Errors: Check and update API keys or credentials
  • Service Connectivity Issues: Verify endpoints, network access, service status
  • Configuration Problems: Review and correct config files or environment variables
  • Permission Errors: Adjust file permissions or access controls
  • Dependency Issues: Install or update required packages or libraries

Safety Guidelines

  • NEVER proceed with repairs without explicit user approval
  • Always explain what will be changed before making changes
  • Prefer non-destructive solutions when possible
  • Back up or note current state before making changes
  • If uncertain, ask for clarification rather than guessing

Example Usage

User: "The image generation service is failing with a NoneType error"

Agent Phase 1 Response: "I've analyzed the error. The issue appears to be that the image generation API is not properly configured or the API key is missing. The error 'NoneType object is not subscriptable' suggests the service is returning None instead of expected data.

Proposed solution: Check the environment variables or configuration file for the image generation API key and endpoint. We may need to add or update the API credentials.

Shall I proceed with checking the configuration files?"

User: "Yes, proceed"

Agent Phase 2: Executes the approved repair actions and reports results.

Resources

This skill includes example resource directories:

scripts/

Executable code (Python/Bash/etc.) that can be run directly.

references/

Documentation and reference material to be loaded into context as needed.

assets/

Files used within the output Claude produces (templates, images, fonts, etc.).

Any unneeded directories can be deleted.

Extended Example with Discussion Loop

User: "The image generation service is failing with a NoneType error"

Agent Initial Analysis: "I've analyzed the error. The issue appears to be that the image generation API is not properly configured or the API key is missing. The error 'NoneType object is not subscriptable' suggests the service is returning None instead of expected data.

Proposed solutions:

  1. Check environment variables for missing API credentials
  2. Verify the image generation service configuration file
  3. Test the API endpoint directly to confirm it's accessible

What would you like to know more about, or would you like me to proceed with one of these approaches?"

User: "What are the risks of option one?"

Agent: "Checking environment variables is read-only and has no risk. However, if we need to add or update credentials, we would be modifying the environment configuration. The main risk is if we set incorrect credentials, the service might be locked out temporarily. I can show you the current values first before making any changes."

User: "Can you check what's currently configured first?"

Agent: "Absolutely. I will examine the current configuration and show you what's set. This is just a read operation, no changes yet."

User: "Okay, proceed with that"

Agent Phase 2: Executes the approved investigation, reports findings, and awaits further instruction or approval for actual repairs.

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

[00] TnR ABSOLUTE RIGOR — Systemic Verification Doctrine. Forced cognitive exertion for high-stakes investigations. Mandates absolute deconstruction, mathematical proof chains, and ≥95% derived certainty. Zero-tolerance for heuristics or sentiment-based reasoning. Shortcuts are defined as intentional hallucinations.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

diegosouzapw/awesome-omni-skill642026년 3월 2일 업데이트

2agent

무료

Configures 2-Agent workflow between PM and implementation roles. Use when user mentions 2-Agent, 2エージェント, PM連携設定, Cursor設定, Cursor連携, 2-agent運用. Do NOT load for: 単独運用, ワークフロー実行, ハンドオフ処理.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

diegosouzapw/awesome-omni-skill642026년 3월 2일 업데이트

Appends relevant hashtags to journal entries based on content analysis. Intended for LLM execution.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

diegosouzapw/awesome-omni-skill642026년 3월 2일 업데이트

add-agent

무료

Add a new AI coding agent to Agent Sandbox. Creates all required files (Dockerfile, templates, CI, docs) and wires the agent into the CLI, proxy, and build system.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

diegosouzapw/awesome-omni-skill642026년 3월 2일 업데이트

Add a new property to the AI agents database. Use when the user wants to add, create, or introduce a new column, property, field, or feature to track across all agents in the comparison matrix. Handles all four required steps - database updates, groups.json, table display, and GitHub issue templates.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

diegosouzapw/awesome-omni-skill642026년 3월 2일 업데이트

Add new Agent templates to the meta-agent-skills framework.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

diegosouzapw/awesome-omni-skill642026년 3월 2일 업데이트

diegosouzapw의 스킬 모두 보기

이 스킬의 문제 신고