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/pandysp/claude-plugins/understudynpx skills add pandysp/claude-plugins --skill understudygit clone --depth 1 https://github.com/pandysp/claude-pluginsWhat 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 | $0.00183 | $0.00774 |
| Opus 5 | $0.00092 | $0.00387 |
| Sonnet 5 | $0.00037 | $0.00155 |
| Haiku 4.5 | $0.00018 | $0.00077 |
Grade A, and why
understudy 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Understudy: write as the maintainer
A maintainer rejects foreign-looking code even when it's correct. A patch in the wrong idiom is a cost to them (something to mentally translate, reformat, or turn down), so a contribution that doesn't look like theirs raises the bar to merging it, sometimes past the point they'll bother. Your default makes this worse: a generic, competent style that quietly announces "written by an outsider", the same tell that reads as AI. This skill is the discipline of effacing your own hand and writing as the host, so the work arrives already in their voice and costs nothing to accept.
The bar you're aiming for: a reviewer can't find the seam between your code and theirs.
The method
1. Sample how they actually write
Read the host's own code, comments, tests, and commit messages. Read more than the one thing you're touching. The hand shows in naming, comment density and what they bother to comment at all, where state lives, how errors flow, how config is gated, how tests are shaped, how commits are phrased. You'll default to glancing at the code and writing; the work is reading enough that their conventions, not your habits, are what's loaded when you start.
2. Iterate past your first pass
One reading anchors on your own habits with a thin coat of theirs, and the idioms you'd get wrong are the ones it doesn't reveal. Keep sampling until a pass comes up dry. You should be able to name what each pass turned up, not just say you "matched the style". The pull to call it done early is the failure this skill exists to counter; when in doubt, sample again.
Style, not substance
Effacing your hand is about how the code reads, not about silencing your judgment. Match their style; don't defer on substance. If something looks wrong on the merits (a bug, a risk, a real design problem), raise it as you would anywhere. The skill doesn't tell you whether to match or push back on that; it only asks that your style be theirs, not that your opinions go quiet.
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.
- 2d ago First seen · 38 lines · 183 tokens per session scan A 48733c2a24aa
understudy is a skill published in the GitHub repository pandysp/claude-plugins (5 stars, last pushed 11d ago), licensed MIT. It adds 183 tokens to every session and 774 once invoked, about $0.0009 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.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
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…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.