AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Exam Revision Handbook

skill-mianbaofang-exam-revision-handbook-exam-revision-handbook · by mianbaofang

Create source-backed GCSE, IGCSE, A-Level, and College Board AP revision handbooks from official AQA, Edexcel, CAIE, and College Board sources. Use when acquiring a syllabus, decomposing requirements into atomic teaching points, writing worked examples, planning or reviewing educational visuals, rendering a handbook, reviewing HTML, or exporting an approved PDF. Do not use for unsupported curricu…

↗ Website ↗ Repository
— No reviews yet
0 installs
13 views
0.0% view→install

Install

$ agentstack add skill-mianbaofang-exam-revision-handbook-exam-revision-handbook

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • ✓ Prompt-injection patterns
  • ✓ Secret / credential exfiltration
  • ✓ Dangerous shell & filesystem operations
  • ✓ Untrusted network calls
  • ✓ Known-malicious package signatures

What it can access

  • ✓ Network access No
  • ✓ Filesystem access No
  • ✓ Shell / process execution No
  • ✓ Environment & secrets No
  • ✓ Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-mianbaofang-exam-revision-handbook-exam-revision-handbook)

Reliability & compatibility

✓ Security review passed
0 installs to date
— no reviews yet
● 16d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Exam Revision Handbook? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Exam Revision Handbook

Follow every gate and linked contract.

Load The Contracts

  1. Read [workflow-contract.md](references/workflow-contract.md) in full before responding.
  2. Read [revisionguidespec.md](references/revisionguidespec.md) before changing artifacts.
  3. Read the applicable College Board AP or OxfordAQA provider note.
  4. Run python scripts/doctor.py before execution. Use python scripts/run_runtime.py -- for every runtime command.
  5. Use evals/ only for validation and release readiness.

Workflow

  1. First respond with the complete structured preflight, including the explicit external-image-capability question. Infer nothing and block all later work until every field is valid.
  2. Acquire official evidence only for the confirmed board, qualification, A-Level stage, market, subject, and year.
  3. The LLM maps every official requirement to atomic teaching points before writing. Python may extract, validate, record, render, and block; it may not own meaning, content, visual judgment, or approval.
  4. Decide visuals per final teaching point. Decorative text boxes are not explanatory diagrams.
  5. Use this order: rebuild manifest, import assets, record per-visual LLM approval, then render only. Rebuilds invalidate approvals.
  6. Render HTML first. The LLM inspects all of it, records item evidence, repairs, rerenders, and repeats until the exact HTML/render/ledger hashes are approved. Machines cannot approve.
  7. Export PDF only from approved HTML and verify all HTML, review, PDF, and delivery hashes.
  8. Review every batch subject independently; never share or sample conclusions.

Output Contract

  • The LLM owns syllabus meaning, teaching content, visual judgment, and approval; Python owns extraction, validation, rendering, recording, and blocking only.
  • Every official requirement, visual, review item, and delivery file remains traceable to current source and artifact hashes.
  • Hand back the approved PDF plus controlled HTML, evidence, review, and delivery artifacts.

Stop Conditions

Stop instead of guessing when preflight is incomplete, an official source is ambiguous, an LLM-owned artifact is missing, any topic or visual is pending/rejected/stale, hashes disagree, HTML review is incomplete, or the packaged runtime is unavailable. Report the exact blocker and next valid action.

Do not publish or claim parity unless every requirement in [migration-acceptance.md](references/migration-acceptance.md) is passed with current evidence.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Write a review

Versions

  • v0.1.0 Imported from the upstream source.

More in Content & Media