local-file-manager

Read, write, append, and list local files in the session's working directory. Use when you need to persist output to disk, read input files, or manipulate file system safely. Supports text files, JSON, CSV, Markdown.

Safety Notice

This item is sourced from the public archived skills repository. Treat as untrusted until reviewed.

Copy this and send it to your AI assistant to learn

Install skill "local-file-manager" with this command: npx skills add 000sonic/local-file-manager-1-0-0

Local File Manager Skill

This skill provides safe file I/O operations within the session's cwd. It is designed for roles that need to store outputs locally (no cloud sync).

Capabilities

  • Read file: Get contents of a text file
  • Write file: Create or overwrite a file
  • Append file: Add content to existing file
  • List files: Directory listing with filtering
  • Delete file: Remove a file (with safety checks)
  • Copy/Move: Simple file operations

When to Use

Role needs to:

  • Save generated code/analysis to disk
  • Read input documents (PDFs, text, etc.)
  • Append logs or results
  • Create output files in Markdown/JSON/CSV

Usage

# Read a file
file-manager --action read --path output.md

# Write content (from stdin or --content)
file-manager --action write --path result.json --content '{"status":"done"}'

# Append to file
file-manager --action append --path log.txt --content "Job completed at $(date)"

# List files in directory
file-manager --action list --dir . --pattern "*.md"

# Create directory
file-manager --action mkdir --dir reports

# Delete file (with confirmation)
file-manager --action delete --path old_file.txt

Safety

  • Sandboxed to cwd: Cannot access files outside session's working directory
  • Protected files: Cannot delete files starting with . or in ../
  • Size limit: Max file size 10MB (configurable)
  • Dry-run support: --dry-run shows what would happen

Integration with Roles

In role config, enable this skill:

plugins:
  allow:
    - local-file-manager
    - doc-parser

Then in the role's system prompt, guide usage:

When you finish analysis, write the result to a file:
  file-manager --action write --path summary.md --content "$YOUR_MARKDOWN"

Examples

Researcher saving analysis:

file-manager --action write --path analysis_$(date +%Y%m%d).md \
  --content "# Analysis\n\n## Summary\n..." 

Developer saving code:

file-manager --action write --path src/main.py --content "$CODE"

Automation appending log:

file-manager --action append --path /var/log/automation.log \
  --content "[$(date)] Task completed\n"

Error Handling

  • If file doesn't exist for read: returns error code 1
  • If path is outside cwd: denied
  • If write fails (permission): returns error
  • All errors logged to ~/.openclaw/logs/file-manager.log

Configuration

Environment variables:

  • FILE_MANAGER_MAX_SIZE: Max file size in bytes (default 10485760)
  • FILE_MANAGER_LOG: Path to operation log (default ~/.openclaw/logs/file-manager.log)
  • FILE_MANAGER_DRY_RUN: Set to "1" to only simulate operations

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

respiratory_symptom_recognition_analysis

Based on computer vision, automatically detects coughing, phlegm, and wheezing frequency, counts the frequency of episodes, used for early health anomaly alerts, helping to detect respiratory diseases in a timely manner. | 呼吸道症状智能识别技能,基于计算机视觉自动检测咳嗽、咳痰、喘息频率,统计发作频次,用于健康异常早期提醒,帮助及时发现呼吸道疾病

Archived SourceRecently Updated
General

stroke-risk-screening-analysis

Combines TCM facial feature recognition with physiological indicator information to provide early warnings of high-risk stroke conditions such as cerebral infarction and cerebral hemorrhage, and provides lifestyle intervention suggestions and medical guidance. | 脑卒中风险筛查技能,结合中医面部特征辨识结合生理指标信息,提前预警脑梗塞、脑出血等脑卒中高危状态,给出生活干预建议和就医指引

Archived SourceRecently Updated
General

stranger-recognition-analysis

Identifies strangers appearing in surveillance areas through facial comparison; supports video stream and image detection, suitable for stranger warnings in residential communities, units, access control, and other scenarios. | 陌生人识别技能,通过人脸比对识别监控区域出现的陌生人员,支持视频流和图片检测,适用于小区、单位、门禁等场景的陌生人预警

Archived SourceRecently Updated
General

layered-memory-sys

分层记忆系统(Layered Memory System)— 为AI助手提供类人的6层记忆架构。 自动管理记忆的生命周期:闪存(3天)、活跃(7天)、关注(30天)、沉淀(90天)、核心(永久)。 包含梦境模式(Dream Cycle)自动执行记忆巩固、归档、遗忘和合并。 支持语义相似度检测和Session日志优先检索。 当用户需要:(1) 优化AI记忆管理,(2) 建立分层TTL记忆系统,(3) 实现自动记忆整理, (4) 搭建梦境模式处理流程,(5) 改善AI跨会话记忆连续性 时使用。 关键词:记忆、记忆系统、分层记忆、TTL、梦境模式、记忆归档、语义相似度、session日志

Archived SourceRecently Updated