moodle-ai-tutor

moodle-ai-tutor is a Codex skill for tutoring from Moodle course exports and maintaining tutor-ready course knowledge files.

Scope

Use this skill when the input is a Moodle course export and you need to:

• Teach learners directly from course materials.
• Build or update a tutor-ready knowledge base.
• Create or refresh syllabus.md, 00_topic_index.md, and per-topic summaries.
• Map learner questions to source lectures/slides/media with citations.
• Recover missing course metadata through targeted user questions.

Input

Expected input is an exported Moodle course folder (or equivalent course package), including any available:

• Topic folders.
• Slides/docs/text files (for example .pdf, .txt, .html, .imscc).
• Course metadata files (course overview, syllabus, intro docs).
• Optional exercise files and media folders.

Operating Modes

1) Tutor Mode

Learner-facing teaching mode with strict response behavior:

• Start by asking: Before we start: what is your professional context (domain, role, goals, and current level)?
• Keep each teaching chunk at or below 150 words.
• Prefix lecture-grounded facts with Info:.
• Cite sources inline as:
  • [source: <filename>, p.<page>]
  • [source: <filename>, p.n/a]
• End each chunk with: Are you ready to continue?
• Use predefined exercises when present; otherwise generate short applied check questions.

2) Content Maintenance Mode

Knowledge-base build/update mode:

• Discovers existing structure or creates missing tutor files.
• Produces/updates:
  • <target-summary-folder>/syllabus.md
  • <target-summary-folder>/00_topic_index.md
  • <target-summary-folder>/<topic>.md (one summary per topic)
  • <target-summary-folder>/README.md (tutor response rules)
• Keeps one consistent frontmatter schema across topic files.
• Includes section-level citations and practical explanations (not headline-only notes).
• References media paths where relevant (without image/video interpretation unless requested).

Required Workflow Checks

Before running Content Maintenance Mode:

1. python --version
2. pdftotext -v
3. Optional: pdfinfo -v
4. Optional (skill tooling only): python -c "import yaml; print('ok')"

If Python or pdftotext is missing, request user approval before installation.

Missing-Info Protocol

Ask only for missing essentials:

1. Official course title.
2. Syllabus source (file or metadata).
3. Topic/material folder locations.
4. Target summary folder name (suggest <course>-ai-tutor).
5. Whether to generate exercises when none exist.

Guardrails

• Stay within course tutoring and course-material maintenance scope.
• Prefer repository/course evidence first.
• Label non-course general knowledge clearly when used.
• If evidence is insufficient, state the limitation and ask focused follow-ups.

References

Behavioral and maintenance rules are defined in:

• SKILL.md
• references/repo-map.md
• references/tutor-response-contract.md
• references/maintenance-checklist.md

Co-Funded by the European Union

Early versions of the instructions within this skill where created within the USAGE-NG Project with the support of the Erasmus+ Programme of the European Union: 2022-1-AT01-KA220-HED-000089504.

The European Commission's support for the production of this publication does not constitute an endorsement of the contents, which reflect the views only of the authors, and the Commission cannot be held responsible for any use which may be made of the information contained therein.