By name: feedback description: Review writer submissions using the Writing Mentor Framework user-invocable: true
Review Submissions
Review all submissions in this folder using the Writing Mentor Framework.
What This Does
- Checks for required dependencies (prompts to install if missing)
- Extracts text from all submissions (DOCX, PDF, XLSX)
- Renders Excel charts for visual review
- Reviews each submission with isolated context (prevents bias)
- Writes two-tier feedback (Reviewer Notes + Writer Feedback)
Prerequisites
Before running, ensure you have:
assignment.md- Your assignment requirementsrubric.md- Your evaluation criteriasubmissions/- Folder containing writer work
Optional:
course_concepts.md- Domain concepts for assumption validationturnitin/- Similarity reports
Workflow
Check dependencies by running:
python {framework_path}/skills/feedback/scripts/check_dependencies.pyIf dependencies are missing, prompt the user to install them before continuing.
Verify required files exist:
assignment.md(required)rubric.md(required)submissions/folder with at least one file (required)
If any are missing, tell the user what's needed and stop.
Extract text from submissions:
python {framework_path}/skills/feedback/scripts/extract_submission_text.py --input submissions --out feedback_extractedRender Excel charts (if any .xlsx files exist):
python {framework_path}/skills/feedback/scripts/render_xlsx_quicklook.py --input submissions --out feedback_renderedRead the framework instructions from
{framework_path}/skills/feedback/SKILL.mdRead reference materials:
assignment.mdrubric.md{framework_path}/skills/feedback/references/economical_writing_principles.mdcourse_concepts.md(if present)
Enumerate submissions by parsing filenames. Group by username.
Read parallelism setting from
wmf-config.yaml:review.max_parallel_agents(default: 3)Process submissions in parallel batches:
- For each batch of N submissions (where N = max_parallel_agents):
- Spawn N Task agents in a single message with
subagent_type: "general-purpose" - Each agent follows the isolated review workflow from the framework SKILL.md
- Wait for all N to complete before starting the next batch
- Spawn N Task agents in a single message with
- Each agent writes to
feedback/{username}.md
- For each batch of N submissions (where N = max_parallel_agents):
Report completion with summary of submissions reviewed.
Parallelism Settings
Configure review.max_parallel_agents in wmf-config.yaml:
| Setting | Use Case |
|---|---|
1 |
Sequential mode — use if you experience rate limits or errors |
3 |
Default — good balance of speed and reliability |
5-10 |
Large classes with hundreds of submissions |
Multi-Round Reviews
If submissions are organized in rounds (submissions/round1/, submissions/round2/):
- Check
wmf-config.yamlforsubmissions.rounds.enabled: true - Use round-specific folders:
feedback_extracted_round{N}/,feedback_round{N}/ - For resubmissions, compare to prior round feedback per config setting
Output
Feedback files are written to feedback/ (or feedback_round{N}/ for rounds).
Each file contains:
- Section A: Reviewer Notes - Technical audit for instructor only
- Section B: Writer Feedback - Teaching-focused guidance to share with writer