Hermes Work Visualization

将Hermes Agent的工作过程可视化 - 包括任务进度、技能使用、代码改动、会话统计等

Security Scan

The skill's code and runtime instructions are consistent with a local work-visualization/monitoring utility that reads local Hermes skill metadata and writes reports to ~/.hermes; there are no obvious network exfiltration or unrelated credential requests, though there are some minor inconsistencies and behaviors you should review before running.

Purpose & Capability

The name/description (visualize Hermes agent work, reports, skill usage, code changes) matches what the files implement: report generation, session summaries, skill statistics and a monitor. The scripts read local files under ~/.hermes/skills/... and produce output under ~/.hermes/skills/work-visualization/output which is coherent with the stated purpose.

Instruction Scope

SKILL.md instructs you to run local Python scripts under ~/.hermes/skills/work-visualization/scripts/*.py. Those scripts: read config and i18n files, enumerate other skill directories (reads _meta.json from ~/.hermes/skills), read /proc/meminfo (monitor), write log/reports to ~/.hermes/skills/work-visualization/{output,cache,cache/monitor.log}, and call pgrep/kill when stopping the monitor. All of this is plausible for a monitoring/analytics tool, but you should be aware they access other skill metadata and can kill processes that match a pattern.

Install Mechanism

There is no install spec (instruction-only for runtime), and all code is provided in the package. No external downloads, package registry pulls, or archive extraction are present in the manifest. Risk from installation mechanism is minimal.

Credentials

The skill requests no secrets or special env vars. It uses language-related env vars (HERMES_LANG, LANG, LC_ALL) for localization, which matches the i18n feature. No unexpected credentials, API keys, or unrelated environment variables are required.

Persistence & Privilege

The skill writes logs, cache and generated reports to ~/.hermes/skills/work-visualization and can spawn a long-running monitor process. It is not always:true and does not alter other skills' configurations. However, stop_monitor uses pgrep/kill on processes matching 'work-visualization.*monitor.py' which could terminate any process whose command line matches that pattern; review that behavior if you have similarly-named processes.

Assessment

This package appears to be a local reporting/monitoring tool and does what it says: generate session summaries, reports and skill usage stats. Before running or enabling it, consider: 1) Review the scripts locally — they will read and write files under your home directory (~/.hermes/skills/work-visualization) and read other skills' _meta.json files to build statistics. 2) The monitor reads /proc/meminfo and writes monitor.log to the cache directory; it can run indefinitely as a background process. 3) The stop command uses pgrep/kill to terminate matching processes — verify the pattern won't accidentally match other important processes. 4) There are duplicate/inconsistent i18n helper modules and some truncated/stubbed code fragments visible in the package (potential runtime errors); test in a safe environment first (or inspect/fix code) before using in production. 5) If you are concerned about privacy, inspect the generated outputs and the content of _meta.json files to ensure they don't leak anything you consider sensitive. If everything looks acceptable, run the scripts from a user account where writing to ~/.hermes is expected and safe.

工作可视化 | Work Visualization

将Hermes Agent的工作过程和结果进行可视化展示，帮助你更好地了解、追踪和分析AI的工作情况。

Visualize Hermes Agent's work process and results to help you better understand, track, and analyze AI performance.

---

语言设置 | Language Settings

中文说明

本技能支持自动语言检测，会根据您的系统语言环境自动选择中文或英文。

手动设置语言：

# 设置为中文
export HERMES_LANG=zh

# 设置为英文
export HERMES_LANG=en

English Instructions

This skill supports automatic language detection and will automatically select Chinese or English based on your system language.

Manually set language:

# Set to Chinese
export HERMES_LANG=zh

# Set to English
export HERMES_LANG=en

---

适用场景 | Use Cases

中文

• 任务进度追踪 - 实时查看当前任务完成情况
• 工作统计分析 - 了解技能使用频率、工具调用次数等
• 代码改动追踪 - 可视化代码编辑历史和影响范围
• 会话总结 - 自动生成工作摘要和关键指标
• 性能监控 - 监控AI工作的效率和资源使用

English

• Task Progress Tracking - View current task completion status in real-time
• Work Statistics Analysis - Understand skill usage frequency, tool call counts, etc.
• Code Change Tracking - Visualize code editing history and impact scope
• Session Summary - Automatically generate work summaries and key metrics
• Performance Monitoring - Monitor AI work efficiency and resource usage

---

使用方法 | Usage

生成工作摘要 | Generate Session Summary

中文

当完成一项重要工作后，运行以下命令生成可视化摘要：

# 生成当前会话的工作摘要
python3 ~/.hermes/skills/work-visualization/scripts/session_summary.py

# 生成特定时间范围的工作报告
python3 ~/.hermes/skills/work-visualization/scripts/generate_report.py --days 7

# 查看技能使用统计
python3 ~/.hermes/skills/work-visualization/scripts/skill_stats.py

English

After completing important work, run the following command to generate a visual summary:

# Generate current session summary
python3 ~/.hermes/skills/work-visualization/scripts/session_summary.py

# Generate work report for specific time range
python3 ~/.hermes/skills/work-visualization/scripts/generate_report.py --days 7

# View skill usage statistics
python3 ~/.hermes/skills/work-visualization/scripts/skill_stats.py

---

可视化类型 | Visualization Types

1. 任务进度可视化 | Task Progress Visualization

中文

• 显示当前待办任务列表
• 任务完成状态和进度条
• 任务优先级和时间预估
• 任务依赖关系图

English

• Display current todo task list
• Task completion status and progress bar
• Task priority and time estimation
• Task dependency graph

2. 技能使用分析 | Skill Usage Analysis

中文

• 最常用技能排行榜
• 技能调用趋势图
• 技能成功率统计
• 技能组合使用模式

English

• Most used skills ranking
• Skill call trend chart
• Skill success rate statistics
• Skill combination usage patterns

3. 工具调用统计 | Tool Call Statistics

中文

• 工具使用频率排行
• 工具调用时间分布
• 工具失败率和错误类型
• 工具执行耗时分析

English

• Tool usage frequency ranking
• Tool call time distribution
• Tool failure rate and error types
• Tool execution time analysis

4. 代码改动追踪 | Code Change Tracking

中文

• 文件编辑历史时间线
• 代码改动统计（增删行数）
• 文件修改频率热力图
• 改动影响范围分析

English

• File editing history timeline
• Code change statistics (added/deleted lines)
• File modification frequency heatmap
• Change impact scope analysis

5. 会话概览 | Session Overview

中文

• 会话关键指标卡片
• 工作时间轴
• 重要决策节点
• 输出成果摘要

English

• Session key metrics cards
• Work timeline
• Important decision nodes
• Output summary

---

可视化输出示例 | Visualization Output Examples

任务进度图 | Task Progress Chart

中文

[===================== 75%]  修复登录问题 (3/4 子任务完成)
[=============------ 65%]  优化数据库查询 (13/20 子任务完成)
[======-------------- 30%]  编写API文档 (3/10 子任务完成)

English

[===================== 75%]  Fix login issues (3/4 subtasks completed)
[=============------ 65%]  Optimize database queries (13/20 subtasks completed)
[======-------------- 30%]  Write API documentation (3/10 subtasks completed)

---

配置选项 | Configuration Options

中文

在 ~/.hermes/skills/work-visualization/config.json 中配置：

English

Configure in ~/.hermes/skills/work-visualization/config.json:

{
  "enabled_metrics": [
    "task_progress",
    "skill_usage",
    "tool_calls",
    "code_changes",
    "session_summary"
  ],
  "report_frequency": "daily",
  "export_formats": ["html", "markdown"],
  "charts": {
    "type": "ascii",
    "show_legend": true,
    "max_items": 10
  }
}

---

语言配置 | Language Configuration

自动检测 | Auto Detection (默认/Default)

系统会自动检测以下环境变量：

• HERMES_LANG - 优先级最高
• LANG - 系统语言环境
• 其他语言相关的环境变量

System automatically detects the following environment variables:

• HERMES_LANG - Highest priority
• LANG - System language environment
• Other language-related environment variables

支持的语言 | Supported Languages

语言代码	语言	Language Code	Language
zh	中文	zh	Chinese
en	英文	en	English

---

扩展建议 | Extension Suggestions

中文

• 集成更多图表类型（饼图、热力图）
• 支持自定义可视化模板
• 添加工作流分析
• 实现跨会话工作追踪
• 支持团队协作视图

English

• Integrate more chart types (pie chart, heatmap)
• Support custom visualization templates
• Add workflow analysis
• Implement cross-session work tracking
• Support team collaboration view

---

相关技能 | Related Skills

中文

• jupyter-live-kernel - 交互式数据分析
• note-taking - 工作笔记记录
• obsidian - 知识库管理
• github-code-review - 代码审查

English

• jupyter-live-kernel - Interactive data analysis
• note-taking - Work note taking
• obsidian - Knowledge base management
• github-code-review - Code review