Syllabus Reader

Parse university course syllabi and extract structured information with proactive risk flagging — hidden pitfalls, grade traps, attendance triggers, and workload collisions that students easily overlook.

Give it a syllabus (PDF, DOCX, URL), get back a comprehensive breakdown. Works on WorkBuddy, Claude Code, Codex, Cursor, Windsurf, Copilot Chat — any AI coding tool.

Works with university syllabi worldwide — covers all major grading systems, term structures, and policy conventions.

---

Installation

One-Line Install (Recommended)

Bash / Git Bash:

curl -fsSL https://raw.githubusercontent.com/JeromeWang6066/syllabus-reader-skill/main/install.sh | bash -s -- <platform>

PowerShell:

irm https://raw.githubusercontent.com/JeromeWang6066/syllabus-reader-skill/main/install.ps1 | iex; .\install.ps1 -Platform <platform>

Replace <platform> with one of:

• workbuddy → ~/.workbuddy/skills/ (git clone, auto-loads)
• claude-code → ~/.claude/rules/ (global rule, all projects)
• codex → ~/.agents/skills/ (git clone, native skill support)
• copilot → ~/.copilot/instructions/ (user-level, all workspaces)
• windsurf → ~/global_rules.md (appends to global rules)
• cursor → ~/.cursor/rules/ (global user rule)

Clone & Run Locally

git clone https://github.com/JeromeWang6066/syllabus-reader-skill.git
cd syllabus-reader-skill

# Bash:
./install.sh <platform>

# PowerShell:
.\install.ps1 -Platform <platform>

# Or install everywhere at once:
./install.sh --all          # Bash
.\install.ps1 -All          # PowerShell

Uninstall

./install.sh --uninstall <platform>
.\install.ps1 -Uninstall <platform>

---

Supported Platforms

Each platform has its own native config directory for global rules/instructions. The installer places the prompt there so it works across all projects automatically — no copying or pasting needed after initial setup.

Platform	Config Directory	Method	Scope
WorkBuddy	~/.workbuddy/skills/	git clone into skills directory	All sessions, auto-loads
OpenAI Codex CLI	~/.agents/skills/	git clone into skills directory	All sessions, /skills to invoke
Claude Code	~/.claude/rules/	Global rule file (*.md)	All projects, auto-discovers
GitHub Copilot Chat	~/.copilot/instructions/	User-level instruction (*.instructions.md)	All VS Code workspaces, auto-discovers
Windsurf	~/global_rules.md	Append to single global rules file	All projects, requires restart
Cursor	~/.cursor/rules/	User rule file (*.mdc)	Global, reference via @ in chat

Note: For platforms without a file-based global config (ChatGPT, Claude web, Gemini), copy prompt.md and paste it as your system prompt or custom instructions.

WorkBuddy

./install.sh workbuddy
# → git clones to ~/.workbuddy/skills/syllabus-reader
# → Restart WorkBuddy → ready to use

The SKILL.md YAML metadata is read automatically. No extra configuration needed.

OpenAI Codex CLI

./install.sh codex
# → git clones to ~/.agents/skills/syllabus-reader
# → In Codex session: type /skills to see and activate

Codex natively supports skills in this directory — same format as WorkBuddy's SKILL.md (YAML frontmatter + Markdown).

Claude Code

./install.sh claude-code
# → writes ~/.claude/rules/syllabus-reader.md (global rule)
# → Available immediately in ALL projects

Claude Code loads ~/.claude/rules/*.md at the start of every session, across all projects. The rule auto-discovers based on context — when you mention a syllabus, it activates.

GitHub Copilot Chat

./install.sh copilot
# → writes ~/.copilot/instructions/syllabus-reader.instructions.md
# → Available in every VS Code workspace automatically

Copilot recursively scans ~/.copilot/instructions/ and applies user-level instructions globally. Uses applyTo: '**' so it activates on all files.

Windsurf

./install.sh windsurf
# → appends prompt content to ~/global_rules.md
# → *** RESTART Windsurf ***

Windsurf reads ~/global_rules.md on startup. Changes require an application restart. The section is cleanly delimited so uninstallation is non-destructive.

Cursor

./install.sh cursor
# → writes ~/.cursor/rules/syllabus-reader.mdc (user rule)
# → Reference @syllabus-reader in chat to invoke

Cursor supports user-level rules via .cursor/rules/*.mdc. Set alwaysApply: false so it only loads when referenced or when relevant.

---

Features

Six Extraction Categories

Category	What It Covers
Basic Info	Course name, code, instructor, contact, credits, prerequisites, semester
Schedule	Lecture/lab times, exam dates, holidays, add/drop deadlines, weekly overview
Teaching Content	Course description, learning objectives, textbooks, weekly topic breakdown
Grading	Component weights, grading scale, late policies, extra credit
Other Notes	Attendance policy, academic integrity, device policy, communication rules
Risk Flags	Proactive scanning for hidden risks and pitfalls (the key differentiator)

Proactive Risk Flagging

Students often skim syllabi and miss critical policies. The prompt performs a dedicated risk scan pass to catch:

Severity	What It Flags
[CRITICAL]	Attendance-based failure triggers, hard prerequisites that can drop you
[HIGH]	Double-pass rules, minimum final exam thresholds, workload collision weeks
[MEDIUM]	Hidden costs (iClickers, access codes), syllabus quizzes, group work risks

Nine risk categories scanned for:

• Attendance Fail Triggers — "3 absences = automatic F"
• Grade Threshold Traps — "Must pass final to pass course"
• Workload Collision Detection — weeks where exam + project + problem set converge
• Late Work Traps — zero-tolerance policies, aggressive hourly deductions
• Hidden Costs — mandatory access codes, iClickers, custom editions
• Syllabus Quiz Requirements — must pass to stay enrolled
• Group Work Risks — peer evaluations that sink your grade
• Ungraded-But-Required — mandatory items with no grade weight
• Strict Prerequisites — enforced vs. advisory "recommended background"

Output Language

• Default: Chinese (Simplified)
• If you chat in English: Output switches to English
• Never translated: course names, book titles, personal names, university names, direct quotations

---

Repository Structure

syllabus-reader-skill/
├── SKILL.md                                  # Canonical instructions (all platforms)
├── prompt.md                                 # Condensed universal prompt
├── install.sh                                # Universal installer (Bash)
├── install.ps1                               # Universal installer (PowerShell)
├── README.md                                 # This file
├── LICENSE                                   # MIT License
├── assets/
│   └── syllabus_summary_template.md          # Output template (Chinese default)
├── references/
│   └── syllabus_patterns.md                  # Pattern library for worldwide syllabi
└── scripts/
    └── generate_ics.py                       # Optional: Generate .ics calendar file

Key Files

File	Purpose
SKILL.md	The canonical instruction set. Works on every platform — ignore the YAML frontmatter if not using WorkBuddy. Contains the most complete version with detailed risk categories, regional adaptation logic, and output formatting rules.
prompt.md	Condensed universal prompt. Self-contained, platform-agnostic. Paste into any AI tool that doesn't have a config directory.
references/syllabus_patterns.md	Comprehensive library of section headers, grading patterns, terminology, and risk indicators covering major university systems worldwide.
assets/syllabus_summary_template.md	Structured Markdown output template. The standard format for results regardless of platform.
scripts/generate_ics.py	Standalone Python utility. Feed a JSON array of events → .ics file for Apple Calendar / Outlook / Google Calendar.

---

Usage Examples

From a URL:

Analyze this syllabus: https://example.com/course-syllabus.html

From a file:

Upload a PDF or DOCX and say:
"请帮我解读这份教学大纲" or "Parse this syllabus for me"

Targeted questions:

"Are there any attendance fail triggers in this syllabus?"
"What are the heaviest workload weeks?"
"Is there a double-pass rule?"
"What hidden costs should I budget for?"

---

Example Output (Abbreviated)

# Design and Analysis of Algorithms — 课程大纲解读

> 快速概览
> - 课程代码：COMPSCI 308  |  学分：4  |  学期：Fall 2026
> - 风险等级：中  |  致命红线：无

## 六、特别关注
### [高危] 重要提醒
| 风险项 | 详情 | 来源 |
|--------|------|------|
| 任务密集周 | Week 9: Midterm + Project Phase 2 截止 | Schedule section |

### 任务密集周
| 周次 | 任务 | 建议 |
|------|------|------|
| Week 9 (Oct 19-23) | Midterm Exam + Project Phase 2 Due | 提前一周开始复习和收尾 |

---

Contributing

Contributions welcome! Found a syllabus pattern that doesn't get detected? Have ideas for new risk categories?

1. Fork the repository
2. Create a feature branch (git checkout -b feature/new-risk-pattern)
3. Add patterns to references/syllabus_patterns.md
4. Update SKILL.md and prompt.md if the detection logic changes
5. Submit a pull request

---

License

MIT — see LICENSE.

---

Acknowledgments

Built from real-world syllabus analysis across university systems worldwide. The risk flagging system was developed by analyzing syllabi from multiple countries and grading schemes to identify the patterns that most frequently catch students off guard.