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/understudylabs/understudy-agent-tools/understand-workloadnpx skills add understudylabs/understudy-agent-tools --skill understand-workloadgit clone --depth 1 https://github.com/understudylabs/understudy-agent-toolsWhat 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 | $0.00075 | $0.02297 |
| Opus 5 | $0.00037 | $0.01149 |
| Sonnet 5 | $0.00015 | $0.00459 |
| Haiku 4.5 | $0.00007 | $0.00230 |
Grade A, and why
understand-workload 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Understand the workload (profile traces, data, prompts, and code path)
The intermediate step before you compare, optimize, or route models. You cannot improve an AI workload you cannot explain. Seeing generated questions or a side-by-side duel in isolation is secondary; first help the user understand what their workload is meant to accomplish, what data it represents, and how a request moves through the app. This skill turns traces, prompt files, datasets, and code paths into a shared mental model — purpose, inputs, outputs, steps, tool-call flow, data shape, failure modes, and success criteria — built with the user through Q&A.
Every workload is different, so this is a skill, not a script: the agent extracts structure from traces and code; the user confirms the task meaning.
Safety Gates
- Redact customer content. Captured prompts often contain real customer data (transcripts, PII, business records). Show structure — system-prompt outline, message roles + sizes, tool catalog, output schema — never raw message bodies. Build the decomposer to redact by construction (sizes and headings, not content).
- Local-first. The decomposition/understanding doc stays local; do not commit it or paste customer payloads into external services. The generated questions must be synthetic (no customer data) before they can be committed or sent to a model.
- Make cost/latency/model claims from the capture itself, not memory.
- No premature duel. Do not route to a frontier-vs-local head-to-head until the workload purpose, data shape, and success criteria are clear enough that the comparison questions map to real task behavior.
Inputs it handles
- Understudy capture envelopes (
.jsonlwith acustomer_request_body). - Raw request JSON (Anthropic/OpenAI shape:
model,system,messages,tools). - Prompt/config files embedded in an app (
prompts/, route handlers, agent policy files, YAML/JSON configs, eval manifests). - Code paths that assemble the request, call the provider, parse responses, invoke tools, retry, stream, or write state.
- Datasets and eval rows (
.jsonl, fixtures, golden outputs, trace exports, benchmark tasks, request logs). - A folder of captures — pick one or two representative ones (e.g. median and largest token count) rather than all of them.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 161 lines · 75 tokens per session scan A 92ea1b092db2
understand-workload is a skill published in the GitHub repository understudylabs/understudy-agent-tools (16 stars, last pushed 3d ago), licensed MIT. It adds 75 tokens to every session and 2,297 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 2026-08-30.
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.
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.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
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…