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.
npx skills add nirholas/ai-tutor-mcp --skill tutor-ledgergit clone --depth 1 https://github.com/nirholas/ai-tutor-mcpWrote 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/nirholas/ai-tutor-mcp/tutor-ledger)<a href="https://agentmods.dev/skills/nirholas/ai-tutor-mcp/tutor-ledger"><img src="https://agentmods.dev/badge/skills/nirholas/ai-tutor-mcp/tutor-ledger/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/nirholas/ai-tutor-mcp/tutor-ledger"><img src="https://agentmods.dev/badge/skills/nirholas/ai-tutor-mcp/tutor-ledger.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 | $0.00057 | $0.00589 |
| Opus 5.5 | $0.00023 | $0.00236 |
| Sonnet 5.5 | $0.00011 | $0.00118 |
| Haiku 4.5 | $0.00006 | $0.00059 |
Grade A, and why
tutor-ledger 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 today.
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 — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pay-As-You-Learn tutor ledger
The plugin starts @three-ws/tutor-mcp over stdio (npx -y @three-ws/tutor-mcp, bin tutor-mcp). It needs no key or signer. The questions themselves are billed elsewhere. This server only reads and seals the ledger.
Environment (both optional)
THREE_WS_BASE: API origin, defaulthttps://three.ws.THREE_WS_TIMEOUT_MS: per-request timeout in milliseconds.
Tools
| Tool | Argument | Endpoint |
|---|---|---|
load_session |
sessionId (string, 1-100 characters) |
GET /api/tutor/session?sessionId=... |
close_session |
sessionId (string, 1-100 characters) |
POST /api/tutor/session with { sessionId, action: "end" } |
What comes back
load_session returns status (open or closed), questionCount, lineItems (one per answered question: question text, level, output tokens, cost in atomic units, a $x.xxxxxx string and a timestamp), totalAtomics and totalUsd. A session that is already closed also returns its invoice. An unknown sessionId returns an empty open session.
close_session seals the ledger. It sets status to closed, stamps a SHA-256 attestation over the line items and total, and returns closedAt, lineItems, totalAtomics, totalUsd and attestation. After closing, no more questions can be billed to that session. Closing is free and moves no funds, because the learner was already charged per question.
Worked example
Request: "How much has session calc-101 cost, and give me the invoice."
- Call
load_sessionwith{ "sessionId": "calc-101" }and reportquestionCountandtotalUsd. - Closing is irreversible, so confirm with the user that they want the session sealed.
- Call
close_sessionwith{ "sessionId": "calc-101" }and return theattestationwith the itemizedlineItems.
Notes
- Do not invent a
sessionId. Use the one the user gives you, since the id is how the ledger is addressed. - Closing cannot be undone. Always load first and confirm before closing.
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.
- today First seen · 40 lines · 57 tokens per session scan A f453ecadfee2
tutor-ledger is a skill published in the GitHub repository nirholas/ai-tutor-mcp (9 stars, last pushed today), licensed Apache-2.0. It adds 57 tokens to every session and 589 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-11-11.
tutor-ledger compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| tutor-ledgernirholas/ai-tutor-mcp | Skill | 9 | today | Apache-2.0 | A | 57 | Claude Code |
| fatqat-workbenchxi-zhao/OpenQuantum | Skill | 72 | yesterday | MIT | A | 73 | Codex |
| local-rag-searchnkapila6/mcp-local-rag | Skill | 134 | 2mo ago | MIT | A | 80 | Claude Code, Codex |
| blazemeter-usage-billingBlazemeter/bzm-mcp | Skill | 26 | 15d ago | Apache-2.0 | A | 67 | Claude Code, Codex |
| award-travelAwardTravelFinder/mcp | Skill | 12 | 11d ago | MIT | A | 164 | Claude Code |
| find-my-bookings-and-balancesAwardTravelFinder/mcp | Skill | 12 | 11d ago | MIT | A | 97 | Claude Code |
| cancel-or-change-awardAwardTravelFinder/mcp | Skill | 12 | 11d ago | MIT | A | 69 | Claude Code |