materio
claudeMaterio: materials research interviews, supplied-corpus synthesis, lab feasibility, task execution and verified paper handoff.
6 skills
10
claude
hpc
HPC pack: Slurm job arrays and sweeps through the clio-kit Slurm server, Lmod and Spack software environments, parallel HDF5 and MPI-IO, Darshan I/O analysis, and ChronoLog run records. The skills drive the cluster's own command-line tools; the clio-kit Slurm server needs one online launch to build its environment before it runs offline.
/
Plugin hpc from marketplace clio-coder in iowarp/clio-coder. Commands read from the repo's manifest — today.
What it installs
Skills (5)
hpc-chronologUses ChronoLog as durable, time-ordered memory for HPC runs through its MCP server: chrono-mcp from a released ChronoLog 3.x tree, or chronolog-mcp from an unreleased ChronoLog 4.0 development build. Records run events and reads them back, quoting what the server says about completeness. Not for dep
hpc-darshanCharacterizes an application's I/O from a Darshan log: runs darshan-parser through a bundled summarizer, reads volumes, access sizes, sequentiality and collective use, and names the bottleneck the counters support. Not for instrumenting a build or for tuning without a log.
hpc-modulesFinds and loads software on an HPC system through Lmod environment modules, and finds Spack-installed packages, from bash. Not for installing software or for writing the job script's science.
hpc-parallel-ioWrites or reviews parallel HDF5 and MPI-IO code so many ranks share one file correctly and collectively, and checks the result with Darshan counters. Not for serial file reading; not for choosing the science stored in the file.
hpc-sweepRuns a Slurm parameter sweep as an experiment: plans a job array and shows a dry run, submits on the operator's word, checks each run with its declared verify check, and tabulates every run, failed and unverified included. Not for a single job.
More from iowarp/clio-coder
All 39 plugins →life-science
claudeLife-science pack: skills that query the RCSB Protein Data Bank, UniProtKB and PubChem over their public, key-free REST APIs with web_fetch, and stamp each answer with the source's entry revision or release. Every lookup needs network; offline the skills say the source is unreachable.
3 skills
10
literature
claudeLiterature pack: verifies citations and resolves DOIs and arXiv IDs through the CiteNexus MCP server (cite-nexus-mcp 0.2.2, key-free Crossref, DataCite, Europe PMC and arXiv), with an offline extractor for BibTeX and Markdown and a web_fetch fallback. Adds a hypothesis workflow: cited grounding, independent council proposals, a testability ranking and the smallest experiment per hypothesis, ending at the operator's own choice. Extends arxiv-literature. Needs network for every lookup; the server needs one online launch to install.
2 skills
10
architecture
claudeDecides the engineering approach for an intent in an interactive session: investigates, proposes two or three genuinely different approaches with trade-offs, recommends, lets the user decide, and writes the decision doc. Not a task-by-task plan; use sprint-plan. Not a multi-perspective debate; use design-council. Not product intent; use product-intent. Not a typed implementation handoff; use tech-spec.
1 skill
10
arxiv-literature
claudeSearches arXiv, summarizes or compares papers, and builds a compact literature survey as citation-ready, source-linked paper cards, with BibTeX export and ID verification through the CiteNexus server when the literature pack is installed. Prefer the Researcher shadow agent for noisy multi-paper retrieval.
1 skill
10
ast-grep
claudeStructural code search with ast-grep rules: finds code by shape when grep on text is too noisy, and never rewrites it. Not for plain string or filename lookups; use grep or find.
1 skill
10
Maintain hpc?
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.
claude & codex plugins39 plugins