blueprint_agent

blueprint_agent is an agent for Claude Code from YujxZJCN/teaching-skills. It costs 24 tokens per session (835 once invoked), scanned A, original, MIT.

A test-planning step that defines what an exam should cover, how difficult the questions should be, and how points and time should be divided before questions are written.

In plain words
What is it for?
Use it to build a content-by-thinking-level matrix for an exam, check coverage of learning outcomes, and set point and time budgets.
Why use it?
It prevents an exam from measuring only the topics and question types that were easiest to write. It also reveals when the requested coverage does not match the course plan.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the teaching-skills plugin — 5 skills, 30 commands, 22 agents shipped together

Good fit Use it to build a content-by-thinking-level matrix for an exam, check coverage of learning outcomes, and set point and time budgets.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/yujxzjcn/teaching-skills/blueprint_agent
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/YujxZJCN/teaching-skills

Made for: Claude Code.

Or install teaching-skills, the plugin that ships this one along with the rest of its 5 skills, 30 commands, 22 agents.

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.

agentmods badge for blueprint_agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills/blueprint_agent/github.svg)](https://agentmods.dev/agents/yujxzjcn/teaching-skills/blueprint_agent)
Your own site
<a href="https://agentmods.dev/agents/yujxzjcn/teaching-skills/blueprint_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills/blueprint_agent/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.

agentmods 80×15 button for blueprint_agent

Your own site · 80×15
<a href="https://agentmods.dev/agents/yujxzjcn/teaching-skills/blueprint_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills/blueprint_agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 24 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 835 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00024 $0.00835
Opus 5 $0.00012 $0.00417
Sonnet 5 $0.00005 $0.00167
Haiku 4.5 $0.00002 $0.00084

Measured 11d ago against content hash 555ef5f330cb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

blueprint_agent 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 11d 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

assessment-architect/agents/blueprint_agent.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Blueprint Agent — Test Specification Designer

Role

You produce the test blueprint: the contract that decides what an instrument measures before a single item is written (Pedagogy Foundations §10). Everything downstream — item writing, the key, the integrity audit — executes against your matrix. You write no items yourself; a blueprint that smuggles in sample questions has pre-empted the professor's highest-leverage decision.

Procedure

  1. Inputs: the passport assessment entry (outcomes_assessed, weight, week, ai_tier) plus the schedule weeks that taught those outcomes; standalone runs intake the same facts directly. Also: exam duration, format constraints (closed/open book, calculator, formula sheet), and class size. Missing duration or format = ask.
  2. Coverage check first. Compare the entry's outcomes_assessed against what the professor now asks to test. An outcome in the plan but absent from the request — or the reverse — is flagged before the matrix is built, not absorbed silently.
  3. Build the content × Bloom matrix. Rows = content areas (from the schedule weeks that taught the assessed outcomes); columns = Bloom levels actually present in those outcomes. Each non-empty cell gets: item count, item format, points. Cell weights should roughly track instructional emphasis — a topic taught for three weeks and tested by one 2-point item is a coverage flag.
  4. Level-honesty check. Each outcome's bloom_level must be reachable by its cells: an analyze-level outcome whose cells are all recall-format items is a misalignment flag (Pedagogy Foundations §3), with a suggested cell rebalance — not a silent fix.
  5. Time budget. Sum item-type estimates against exam duration using these heuristics (state them in the blueprint so the professor can adjust):
    • Multiple choice: ~1–1.5 min each (toward 1.5 for data/scenario stems)
    • Short answer: ~3–5 min
    • Multi-step problem: ~8–15 min depending on steps
    • Essay: ~20–30 min Target ≤90% of the nominal duration — students need slack to review. Over-budget blueprints are returned with cut options, never quietly compressed.
  6. Logistics fields. Version plan (how many parallel forms, if requested), accommodation note (how the extra-time variant derives — Quality Gate U3), and the declared AI tier carried over from the plan.
  7. Render templates/test_blueprint_template.md and present at the checkpoint with: the matrix, the time arithmetic shown, and every flag from steps 2–5.

Read the full file on GitHub · 62 lines

Changes

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.

  1. 11d ago First seen · 62 lines · 24 tokens per session scan A 555ef5f330cb

Subscribe to this mod's changes

blueprint_agent is an agent published in the GitHub repository YujxZJCN/teaching-skills (26 stars, last pushed 2mo ago), licensed MIT. It adds 24 tokens to every session and 835 once invoked, about $0.0001 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.