cs-phd-thesis-evaluationExpert examiner for computer science PhD theses and doctoral dissertations. Runs a structured, calibrated evaluation — thesis argument reconstruction, a claims-to-evidence ledger, and checks on originality, technical correctness, methodological rigor, reproducibility, significance, coherence, the ca
design-science-researchDesign Science Research (DSR) expert for planning, conducting, writing, evaluating, and reviewing research that builds and evaluates artifacts (constructs, models, methods, instantiations, design principles, and design theories). Covers Hevner's 7 guidelines and 3 cycles, the Peffers DSRM, Action De
openresearch-cliUse the `orx` CLI to run local OpenResearch projects from a terminal — create experiment branches, launch and supervise compute, inspect logs and evidence, and manage the experiment tree. Read this before driving `orx` programmatically.
orx-agent-delegationDelegate independent work to helper agent sessions with `orx agent spawn`: task selection, self-contained briefs, branch ownership, compute authorization, wakeups, and nesting or concurrency constraints. Use before spawning a helper or interpreting its result; do not delegate the literature retrieva
orx-computeLaunch and monitor experiment runs and route guidance for hf, modal, k8s/Kubernetes, ssh, slurm, ray, OpenResearch, Tinker, and local backends. Covers the fixed run contract, sizing, cancellation, and wait versus wake. Use before any launch or relaunch, when authoring a k8s manifest, choosing or swi
orx-createInitialize a project with `orx up` and add experiment nodes with `orx create-experiment`. Use when starting a project or experiment, when the tree is empty, or when choosing a baseline, parent, or run command.
orx-customizeSave reusable skills and LaTeX templates with `orx skills add` and `orx templates add`. Use when the user asks to create, import, or add a skill or template to OpenResearch for use across projects.
orx-evidencePrepare and inspect experiment run evidence: design stdout metrics and summaries, read persisted results with `orx logs`, and validate run-derived claims. Use before launching a run whose output must be judged, after a run finishes, or before analyzing or reporting run results.
orx-experiment-treePlan and drive the experiment tree: first-launch setup, fixed run contract, frozen nodes, stacked-bush tree shape, branch/launch/wait/promote, repair limits, notes, and turn summaries. Use before creating or changing experiments, launching a first run, deciding what to try next, handling a completed
orx-figuresPublication-quality figures in matplotlib or TikZ: learning curves, scaling laws, benchmark and ablation comparisons, Pareto trade-offs, heatmaps and confusion matrices, method diagrams. Covers the shared style module, sizing, uncertainty, and vector export. Use whenever you plot, chart, or visualiz
orx-gitRead, edit, commit, and diff experiment code; coordinate shared worktrees and preserve frozen branch history. Use before branch work, when starting work alongside other sessions, comparing nodes, preparing a run, repairing a provisional node, or diagnosing stale code.
orx-instancesCreate standalone OpenResearch compute instances with `orx instance create`. Use when the user wants a persistent machine for manual or ad-hoc work rather than an experiment run.
orx-lit-reviewSearch and read research papers. The main agent calls alphaXiv, OpenAlex, and bioRxiv discovery primitives, ranks the combined candidates, and chooses sources for focused follow-ups. Use for literature reviews, related work, prior art, papers, authors, methods, benchmarks, or research claims; never
orx-paperDraft an academic paper or preprint as LaTeX. Create a .tex file in the project working tree, where it renders for the user and compiles to PDF. Use for a paper, preprint, manuscript, arXiv or submission draft, or a section of one; generic reports and result summaries belong to `orx-reports`.
orx-reportsWrite and organize durable outputs in the artifacts directory. Use before creating or organizing artifacts, including reports, summaries, comparisons, figures, and exported data, or when a line of work concludes.