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/openhcsdev/openhcs/use-openhcsnpx skills add OpenHCSDev/openhcs --skill use-openhcsgit clone --depth 1 https://github.com/OpenHCSDev/openhcsWhat 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 | $0.00055 | $0.00611 |
| Opus 5 | $0.00028 | $0.00305 |
| Sonnet 5 | $0.00011 | $0.00122 |
| Haiku 4.5 | $0.00006 | $0.00061 |
Grade A, and why
use-openhcs 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.
What it actually says
Use OpenHCS
- Call
openhcs_health_checkbefore relying on other tools. Report a bootstrap or stale-server failure instead of working around it. - If you do not already know OpenHCS, call
openhcs_get_authoring_contextwithkind="first_use". Retain its execution, step-pattern,variable_components/group_by, artifact, and CellProfiler mental model while completing the task. - Treat Python pipeline source as one complete
PipelineDocument:pipeline_configplus orderedpipeline_steps. Never send a steps-only fragment, mirror configuration through a second argument, or strip source bindings from the reviewed document. UI-visible and headless routes use these same declarations even though their process and state owners differ. - Call
openhcs_search_capabilitieswith task-relevant workflow, target, or text filters. Its currentsurface_profile, registry-owned workflow metadata, side effects, and security metadata—not a remembered tool list—decide whether to use a UI-visible or exposed headless route. Useopenhcs_list_capabilitiesonly when the complete selected surface is required. - Search knowledge and examples before authoring. For CellProfiler or benchmark work, search the biological task plus
OpenHCS Pythonand retrieve the returned Official30 section withmax_chars=50000. CellProfiler image, object, measurement, relationship, and export semantics lower into the same OpenHCS declarations and runtime. - Inspect real plate data and registered function declarations before authoring. If the microscope is unsupported, use typed pipeline-level
SourceBindingsConfigdeclarations to filter files, extract metadata, name semantic sources, and project a virtual workspace; make each consumingFunctionStepselect those aliases through its step-local source bindings. Never parse filenames inside processing functions. Reflectglobalorpipelineconfiguration withopenhcs_describe_config_schemabefore setting non-obvious fields. Keep filesystem operations inside configured read and write roots. - Validate and compile before execution. Do not infer that source code, UI state, or an earlier validation result implies a current compiled plan.
- Start read-only. Use capability-registry metadata as the authority for mutation and exposure; before mutation, execution, UI actions, viewer launch, network use, or external data exposure, show the target/change, obtain approval, and refresh revision or request tokens.
- Preserve the active ownership route. Discover the GUI bridge before UI tools and apply code/state changes with current tokens. Use headless tools only when their workflow group is exposed. Treat structured errors and recovery hints as authoritative; never bypass path policy, stale-process checks, compile requirements, or bridge authentication.
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 · 17 lines · 55 tokens per session scan A 0d6ee760ec69
use-openhcs is a skill published in the GitHub repository OpenHCSDev/openhcs (4 stars, last pushed 2d ago), licensed MIT. It adds 55 tokens to every session and 611 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
tornadovm-nvidia
NVIDIA-specific TornadoVM workflows on the CUDA backend — profile with Nsight Systems (nsys) and read NVTX ranges, use the hybrid library-task API (cuBLAS/cuBLASLt/cuDNN/cuFFT/cuSPARSE/cuTENSOR/CUTLASS/NCCL native calls from Java), print the generated CUDA-C kernel, and fuzz-test the CUDA-C backend with tornado-fuzz.…
tornadovm-perf-campaign
Run a measured performance/correctness campaign on the TornadoVM runtime — profile first (JFR for host, Nsight Systems for driver), falsify the hypothesis with a standalone probe, A/B against a same-build baseline, then ship one PR per measured claim. Use when asked to make TornadoVM faster, to find where time goes in…
Bio Image Processing Skills Index
Skills for biological image analysis: cell/nucleus segmentation, image restoration, and spatial data processing.
onboard-model
Onboard or periodically reverify and benchmark a Hugging Face model on an exact target GPU platform. Use when asked to add a model recipe, refresh a maintained recipe on a supplied GPU server, benchmark serving, create reproducible experiments and a durable results report, fully qualify and tune the model's Emmy…
reproduce-article-benchmarks
Use this skill when the user asks to re-run an article's benchmarks, reproduce blog post numbers, validate that an article URL still holds, check whether the latest code still performs like a published post, or otherwise compare re-measured Emmy results with published results. It fetches the article, finds its…
tune-kernels
Tune Emmy kernels for a Hugging Face model, traced IR, or golden YAML. Use when asked to tune a model or golden set, seed MCTS with model-proposed knob configurations, compare hybrid proposals against MCTS-only search, diagnose slow or failing kernels, refresh per-GPU goldens, or produce a per-kernel tuning findings…