Getting it into your agent
There is no command for this one: it runs only inside a plugin, and the catalogue could not identify which plugin ships it. The source is linked below.
Wrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/aaronl725/ielts-writing-review-skills/ielts-task2-review)<a href="https://agentmods.dev/skills/aaronl725/ielts-writing-review-skills/ielts-task2-review"><img src="https://agentmods.dev/badge/skills/aaronl725/ielts-writing-review-skills/ielts-task2-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/aaronl725/ielts-writing-review-skills/ielts-task2-review"><img src="https://agentmods.dev/badge/skills/aaronl725/ielts-writing-review-skills/ielts-task2-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00102 | $0.02353 |
| Opus 5 | $0.00051 | $0.01177 |
| Sonnet 5 | $0.00020 | $0.00471 |
| Haiku 4.5 | $0.00010 | $0.00235 |
Grade A, and why
ielts-task2-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
IELTS Task 2 Review
Core Rule
Imitate the teacher's marking style before giving generic IELTS advice. Use the full teacher samples and the official descriptors as the two authorities:
- Read
references/teacher_style.mdbefore reviewing. - Read
references/ielts_task2_band_descriptors.mdbefore scoring. - Use
references/teacher_samples_index.mdto find relevant sample files, then read specific full samples fromreferences/teacher_samples/when needed. - Use
references/review_format.mdfor the output structure.
Workflow
- Identify the Task 2 prompt and the student's essay.
- Normal DOCX input should be an unreviewed student file containing the prompt, optional outline, and the student's essay.
- For a
.docx, first runscripts/extract_task2_input.pyto separate the prompt, optional outline, and student essay. - Preserve
student_essay_paragraph_indicesfrom the extraction result and include it in the review plan. - Treat short outline bullets after the prompt as planning notes, not as the scored essay.
- Existing comments, italic rewrites, feedback, or model essays are not expected in normal use. If they are detected, treat them as accidental leftover material only; do not score, rewrite, or copy them as part of the student's essay.
- For pasted text, separate prompt and essay from headings, blank lines, or explicit labels.
- Check task fit before polishing language.
- If the essay content, position, or logic is significantly off-topic, mark this first in comments.
- Later italic rewrites and the final model essay must redirect the mistaken idea back to the task.
- Select relevant teacher samples before writing comments.
- Use
scripts/find_teacher_sample.pywith the input filename or prompt. - The script normalizes spaces/underscores, so
C17T3 Writing Task 2.docxmaps toC17T3_Writing_Task_2.json. - If the exact file sample is not relevant, use
references/teacher_samples_index.mdand the prompt wording to choose a closer sample.
- Use
- Split the essay into review units.
- Use sentence-level units when problems are local.
- Use paragraph-level units when logic, task response, or coherence needs bigger repair.
- Add teacher-style comments.
- Comments must be short, English, practical, and anchored to specific words or phrases.
- A non-empty comment target must match text inside the selected student essay paragraph; do not fall back to prompt, outline, or the whole wrong paragraph.
- Prefer concrete notes:
informal,repetition,unnatural,could be more concise,Careful with articles,Off-topic,This is vague.
- Add italic rewrites after the relevant original unit.
- Rewrites should be concise, formal, natural, and at a stable Band 7.5 standard.
- Keep rewrites close to the student's intended meaning; avoid making local fixes sound like an over-polished Band 9 sample.
- Score the student's original essay strictly using official Task 2 descriptors.
- Score Task Response, Coherence & Cohesion, Lexical Resource, Grammatical Range & Accuracy, and estimated overall band.
- For this teacher-style educational review, criterion scores and the estimated overall score may use whole or half bands, such as
6,6.5, or7. Use.5when the original essay sits between adjacent whole-band descriptor anchors; do not force criterion scores to integers. - Do this before giving improvement advice.
- Give concise Band 7.5 / 8.0-oriented feedback.
- Separate issues blocking stable Band 7.5 from improvements needed to move toward 8.0.
- Keep this section short, like the teacher samples.
- Write a 250-300 word model essay at a stable Band 8.0 standard.
- Use exactly four paragraphs:
- Introduction: paraphrase the prompt and state the essay's position.
- Body paragraph 1.
- Body paragraph 2.
- Conclusion: begin with
In conclusion.
- Leave one blank line between model essay paragraphs in the output DOCX.
- Align it with the official criteria.
- Preserve the student's main position when it is on-topic and defensible.
- Correct any flawed or off-topic logic from the original draft.
- It must be strong enough for Band 8.0: a clear and well-developed position, relevant and well-extended support, well-managed cohesion, flexible topic vocabulary, and mostly error-free grammar.
- Keep it realistic and teacher-like, not an over-polished Band 9 essay.
- Use exactly four paragraphs:
- Generate a reviewed
.docxby default.
- For input
MyEssay.docx, outputMyEssay(reviewed).docxin the same folder unless the user specifies another path. - For DOCX input, create the reviewed file by copying the original DOCX first.
- Write comments directly into the copied original essay paragraphs; do not create a second copy of the student's essay for comments.
- Anchor comments only inside extracted student essay paragraphs, never inside prompt or outline paragraphs.
- Insert italic rewrites immediately after the matched original essay paragraph.
- Keep the original prompt and student essay in place; do not repeat the task prompt.
- For pasted text, output to
/Users/aaronliang/Desktop/IELTS_Task2_Reviewed_YYYYMMDD_HHMM.docx. - Use Times New Roman for all added review text and comments.
- Use
Cyber Esmeas the Word comment author. - Do not add big visible section headings such as
Task,Reviewed Essay,Score, orModel Essay. - Preserve the original
word/document.xmlroot namespace declarations andmc:Ignorable; do not leave undeclared prefixes such asw14,w15,w16*, orwp14. - Insert a page break before the score and feedback page; keep the score lines and
To Reach Band 7.5 / 8.0together; insert another page break before the model essay. - Never overwrite the original essay unless explicitly requested.
- Clean up temporary byproducts after a successful review.
- Delete temporary review plan files such as
review_plan_c17t3.json. - Prefer creating review plans in
/private/tmpor use--cleanup-planwhen runningscripts/create_review_docx.py. - Do not delete source essays, final reviewed
.docxfiles, or bundled reference files.
What ships with it
35 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 294 B
- references/ielts_task2_band_descriptors.json 11 KB
- references/ielts_task2_band_descriptors.md 12 KB
- references/review_format.md 5.1 KB
- references/review_plan_example.json 4.0 KB
- references/teacher_samples_index.md 3.0 KB
- references/teacher_samples/C15T1_Writing_Task_2.json 28 KB
- references/teacher_samples/C15T1_Writing_Task_2.md 7.0 KB
- references/teacher_samples/C15T2_Writing_Task_2.json 26 KB
- references/teacher_samples/C15T2_Writing_Task_2.md 7.2 KB
- references/teacher_samples/C15T3_Writing_Task_2.json 29 KB
- references/teacher_samples/C15T3_Writing_Task_2.md 6.8 KB
- references/teacher_samples/C15T4_Writing_Task_2.json 17 KB
- references/teacher_samples/C15T4_Writing_Task_2.md 4.9 KB
- references/teacher_samples/C16T1_Writing_Task_2.json 25 KB
- references/teacher_samples/C16T1_Writing_Task_2.md 6.9 KB
- references/teacher_samples/C16T2_Writing_Task_2.json 24 KB
- references/teacher_samples/C16T2_Writing_Task_2.md 8.5 KB
- references/teacher_samples/C16T3_Writing_Task_2.json 25 KB
- references/teacher_samples/C16T3_Writing_Task_2.md 9.3 KB
- references/teacher_samples/C16T4_Writing_Task_2.json 22 KB
- references/teacher_samples/C16T4_Writing_Task_2.md 7.4 KB
- references/teacher_samples/C17T1_Writing_Task_2.json 18 KB
- references/teacher_samples/C17T1_Writing_Task_2.md 6.4 KB
- references/teacher_samples/C17T2_Writing_Task_2.json 17 KB
- references/teacher_samples/C17T2_Writing_Task_2.md 6.2 KB
- references/teacher_samples/C17T3_Writing_Task_2.json 18 KB
- references/teacher_samples/C17T3_Writing_Task_2.md 6.8 KB
- references/teacher_style.md 3.3 KB
- scripts/create_review_docx.py 34 KB runs code
- scripts/extract_task2_band_descriptors.py 3.7 KB runs code
- scripts/extract_task2_input.py 9.8 KB runs code
- scripts/extract_teacher_samples.py 11 KB runs code
- scripts/find_teacher_sample.py 2.5 KB runs code
- scripts/validate_review_docx.py 9.1 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 149 lines · 102 tokens per session scan A ef0502bba849
ielts-task2-review is a skill published in the GitHub repository AaronL725/ielts-writing-review-skills (31 stars, last pushed 25d ago), licensed MIT. It adds 102 tokens to every session and 2,353 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
frog-and-toad-studio
A lesson-planning studio for teachers who teach English as a second language to first-grade learners using Arnold Lobel's Frog and Toad stories. For each chapter, it creates a projected keyword map, a one-page student workbook, and a workbook answer key.
frog-and-toad-illustration-style
A style guide and prompt library for generating square watercolor illustrations of Frog and Toad course material. It includes ready-made prompt lists for 20 chapters and supports user-supplied prompt files.
english-coaching
Use when responding to any user message in a Claude Code session running the nativish plugin — for non-native English speakers who want inline grammar and spelling feedback alongside every reply.
immersion-reader
Build a local H5 English deep-reading lesson from an article URL, YouTube transcript, podcast transcript, or prepared segments.json.
oral-english-practice
A Claude Code skill that stores and analyses spoken-English practice completed in the Claude app. The app handles the conversation; this skill keeps reports, scores, mistakes, progress trends, and the next practice focus.
lecture-builder
A document-building workflow that turns a teacher's knowledge and problem files into printable student and teacher Word documents.