agentleFS
Runtime-3rr0rRuntime-3rr0r/canvas-scanner/skills/course-knowledge-base/SKILL.md

course-knowledge-base

Skill2 starsChanged 56 days agoMITView on GitHub

Use when maintaining the per-course COURSE-BRAIN index.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Keep one file per course; never a per-week brain.

    line 59
  2. Distinctive:

    Every course folder has a COURSE-BRAIN.md whose last-refreshed date is newer than the most recent sweep/note run.

    line 63
  3. Distinctive:

    Every "Lectures covered" row has a resolvable source link.

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Course Knowledge Base (COURSE-BRAIN)

Maintain a single synthesized knowledge doc per course — <Course>/COURSE-BRAIN.md — an index over the authoritative source files (syllabus, README, notes, materials, announcements) so "ask almost anything about a class" can be answered from one file instead of re-scanning folders.

The index is a pointer, not a store (writing-for-agents): COURSE-BRAIN summarizes + links; it never replaces the source files.

Structure of COURSE-BRAIN.md

# <Course> — COURSE-BRAIN
_Last refreshed: <date> · sources: README, syllabus, Notes/, Announcements/_  ([^1]: ...)

## Snapshot
- Instructor, meeting time/place, grading weights (table), current letter estimate
- Course-specific AI policy + late/attendance policy (one line each)

## Key dates (next 30 days)
| Date | Item | Weight | Where |

## Lectures covered (running index)
| Date | Topic | Notes section | Transcript? |
- append one row per lecture; link `Notes/<file>#<section>` and any transcript/Key Points

## Assignment patterns
- types, cadence, submission channel (e.g. "weekly problem sets, Canvas upload, Fri 11:59 PM")
- the professor's grading quirks (sig figs, show work, no make-ups...)

## Open flags
- study flags with open questions, unresolved to-dos (mirror Roadmap, kept current)
- links out: exam prep maps, formula-sheet provided-vs-memorize notes

## Syllabus change log
- every detected syllabus/policy diff + date (see pitfall)

When to refresh

  1. During the weekly Canvas sweep (after canvas-course-sync finishes) — update key dates, grades, new announcements, policies.
  2. After any study-notes-workflow run that touches the course — new lecture row, flags, mastery changes.
  3. When the user asks a course question — read COURSE-BRAIN first; open linked sources only when the brain doesn't answer.
  4. On syllabus/policy change — diff the current Canvas syllabus vs the stored one; append to the change log instead of silently overwriting.

Submission norms (per course, inside COURSE-BRAIN)

  • Record per course: submission format (PDF/docx/online form), naming convention, size limits, deadline timezone, late policy, and quirks (e.g. "no email submissions, Canvas only").
  • Auto-remind before each submission window ("HW due Fri 11:59 PM — PDF, HW3_<LastName>.pdf, Canvas upload") and catch format errors before upload — see homework-checklist.
  • This block is part of the COURSE-BRAIN structure above (add under "Snapshot" or as its own section).

Pitfalls

  • COURSE-BRAIN is an index: bloating it with full notes defeats the point (see writing-for-agents two-load principle).
  • Refresh AFTER sweeps/note-runs, never before — stale brain is worse than no brain.
  • Grading-policy changes must land in the change log, not silently edited into the snapshot.
  • Keep one file per course; never a per-week brain.

Verification

  • Every course folder has a COURSE-BRAIN.md whose last-refreshed date is newer than the most recent sweep/note run.
  • Every "Lectures covered" row has a resolvable source link.
  • Change log reflects every detected syllabus diff.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.