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/pmndrs/claude-code-plugin/examplesnpx skills add pmndrs/claude-code-plugin --skill examplesgit clone --depth 1 https://github.com/pmndrs/claude-code-pluginWhat 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 | $0.00072 | $0.00988 |
| Opus 5 | $0.00036 | $0.00494 |
| Sonnet 5 | $0.00014 | $0.00198 |
| Haiku 4.5 | $0.00007 | $0.00099 |
Grade A, and why
examples 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
pmndrs examples
The pmndrs MCP server serves the gallery at pmndrs.github.io/examples — 167 demos,
each a small self-contained r3f app with its source. Most r3f questions are "how is
this put together", and for those an example beats a reconstruction: it runs, it is
whole, and someone already made every decision in it.
Reach for the docs (pmndrs:docs) for what an API takes; reach here for what a scene
looks like when it works.
Always index first
ListMcpResourcesToolis not needed — readexamples://indexdirectly withReadMcpResourceTool. It is the whole gallery, ~4.5k tokens, one line per demo.mcp__pmndrs__get_examplewith thenamefrom that line.
Read the index once per conversation. There is no per-tag or per-library variant, and no search tool — the index is the search.
Reading a line
{name} [({title}, when it is not just the name)] · {description} · +{libraries} · #{tags} · ~{size}
Everything after the name is dropped when the demo does not carry it:
aquarium · #transmission
arkanoid · Simple arkanoid implementation using cannon physics. · +cannon,zustand · #physics,game,audio
bounds-and-makedefault (Bounds and makeDefault) · #bounds
flow-shield · Interactive energy shield. · +postprocessing,leva · #shader · ~23k
+ lists what a demo uses on top of @react-three/fiber and @react-three/drei,
which all of them use — so +rapier is a real signal and the absence of + is not.
Do not match on tags alone. They are freeform and unvalidated: expect typos
(gtlf, clell-fracture, frosted-glas) and both spellings of one idea
(contact shadows / contact-shadows). Scan names and descriptions too — 39 of the
167 have no description at all, so a name is sometimes the only thing that says what
a demo is.
How many to open
Most demos are small — the median is ~1.4k tokens and nine tenths are under 4k. Three
of them cost less than the index you have already read, so when two lines match
equally well, open both. That is cheaper than opening one, finding it wrong and coming
back, and it is often the answer outright: comparing caustics against
diamond-refraction is what "transmission or refraction?" is asking for. Same when
nothing does the whole job and the answer is one demo's scroll rig plus another's
material.
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 · 85 lines · 72 tokens per session scan A 326146765246
examples is a skill published in the GitHub repository pmndrs/claude-code-plugin (5 stars, last pushed 18d ago), licensed MIT. It adds 72 tokens to every session and 988 once invoked, about $0.0004 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…