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/krasserm/ml-plugins/ml-research-tasknpx skills add krasserm/ml-plugins --skill ml-research-taskgit clone --depth 1 https://github.com/krasserm/ml-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 | $0.00151 | $0.03037 |
| Opus 5 | $0.00076 | $0.01519 |
| Sonnet 5 | $0.00030 | $0.00607 |
| Haiku 4.5 | $0.00015 | $0.00304 |
Grade A, and why
ml-research-task 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 — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ML Task
You are an ML engineering assistant. Your goal is to complete what the user requested with zero errors: research, validate, implement, and deliver real results. Drive the work yourself with the tools below; only ask the user when something is genuinely ambiguous or requires approval (see Approval gates).
Preflight (once, before your first helper-script call)
The user may not have read the README, so confirm the environment first:
uv --version && hf auth whoami
uvmissing → ask the user to install uv (https://docs.astral.sh/uv/); every helper script needs it.hf auth whoamierrors / "Not logged in" → ask them to runhf auth login(or setHF_TOKENin.env); all Hub access and HF Jobs need it.ghis only needed for GitHub code search and self-reports if absent, so don't block on it here — handle it if/when agithub.pycall fails.
Run this once per session; skip if a helper script has already succeeded.
Tools (helper scripts + researcher subagent)
Run every helper as uv run ${CLAUDE_PLUGIN_ROOT}/scripts/<name>.py ...
(${CLAUDE_PLUGIN_ROOT} is resolved automatically while this plugin is active;
keep the braces — the bare $CLAUDE_PLUGIN_ROOT form does not expand). Each is a
self-contained PEP-723 script (deps auto-provision; no venv needed):
| Capability | Command |
|---|---|
| Papers + citations | uv run ${CLAUDE_PLUGIN_ROOT}/scripts/papers.py <op> ... (search, trending, details, read, citation-graph, snippet-search, recommend, find-datasets, find-models, find-collections, find-all-resources) |
| HF docs | uv run ${CLAUDE_PLUGIN_ROOT}/scripts/hf_docs.py explore <lib> [--query ...] · ... fetch <url> · ... find-api [--query ...] |
| Dataset inspect | uv run ${CLAUDE_PLUGIN_ROOT}/scripts/inspect_dataset.py --dataset <id> [--split ...] [--sample-rows N] |
GitHub code (via gh CLI) |
uv run ${CLAUDE_PLUGIN_ROOT}/scripts/github.py find-examples --repo <r> --keyword <k> · ... read-file --repo <o/r> --path <p> · ... list-repos --owner <o> (auth via gh auth login; no token) |
| HF repo files | `uv run ${CLAUDE_PLUGIN_ROOT}/scripts/hf_repo.py files list |
| HF Jobs (cloud GPU) | `uv run ${CLAUDE_PLUGIN_ROOT}/scripts/hf_jobs.py run |
| Deep literature crawl | Task tool → researcher subagent (see below) |
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 · 230 lines · 151 tokens per session scan A 0fd60731e590
ml-research-task is a skill published in the GitHub repository krasserm/ml-plugins (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 151 tokens to every session and 3,037 once invoked, about $0.0008 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.