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 agentmods add skills/simonschick/code-tutor-narrator/skillnpx skills add SimonSchick/code-tutor-narrator --skill skillgit clone --depth 1 https://github.com/SimonSchick/code-tutor-narratorWhat 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 | $0.00066 | $0.00942 |
| Opus 5 | $0.00033 | $0.00471 |
| Sonnet 5 | $0.00013 | $0.00188 |
| Haiku 4.5 | $0.00007 | $0.00094 |
Grade A, and why
tutor 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 yesterday.
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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tutoring out loud
You drive the user's editor while you talk. The tutor MCP server (from the
Code Tutor Narrator extension) gives you five tools:
show_code— scroll to a range, highlight it, dim the rest, speak the narration. Returns the numbered source lines, so you also read what you just pointed at.walkthrough— a planned sequence of beats, played back-to-back with no gaps.speak— say something without moving the editor.editor_state— active file, cursor, selection, visible lines, current errors.clear_highlight— drop the highlight, stop talking.
If a call fails with connection refused, VS Code isn't open — say so and stop trying; don't fall back to silently pasting code into chat.
A walkthrough that reports a timeout has not failed. It holds its response open
while the narration plays, so a long tour can outlive the MCP request window even
though the audio is still going. Never retry it and never re-narrate those beats —
you would talk over yourself. Carry on as if it succeeded; editor_state will show
which beat it reached.
How to run a session
Start where their attention already is. Call editor_state first. If they
have something selected or their cursor is parked somewhere, begin there rather
than at line 1.
Open with orientation, not code. One speak call saying what the file is for
and where you're going to take them. Two sentences.
Then move in beats. One idea per beat, 5–20 lines each. Never put a whole file into one beat.
Which tool depends on whether you already know what you're going to say:
walkthrough— hand it the planned beats and it plays them continuously, synthesising each while the previous is still speaking. Default to this for a section you've read and planned.show_code— one beat, blocking, returns the source. Use it when you need to read the code back before choosing the next beat, when the user is likely to interrupt, or when you're following their cursor.
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.
- yesterday First seen · 80 lines · 66 tokens per session scan A 2935cd6b732f
tutor is a skill published in the GitHub repository SimonSchick/code-tutor-narrator (0 stars, last pushed 15d ago), licensed MIT. It adds 66 tokens to every session and 942 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…