nix-conf

Search, read, and query research papers via Feynman's alphaXiv-backed alpha tools. Use when the user asks about academic papers, wants to find research on a topic, needs to read a specific paper, ask questions about a paper, inspect a paper's code repository, or manage paper annotations.

Execute research code inside isolated Docker containers for safe replication, experiments, and benchmarks. Use when the user selects Docker as the execution environment or asks to run code safely, in isolation, or in a sandbox.

Explain research, papers, or technical ideas in plain English with minimal jargon, concrete analogies, and clear takeaways. Use when the user says "ELI5 this", asks for a simple explanation of a paper or research result, wants jargon removed, or asks what something technically dense actually means.

Run explicitly chosen research benchmark or replication jobs on Modal's serverless infrastructure. Use when a Feynman research workflow needs burst remote GPU compute and the Modal CLI is available.

Preview Markdown, LaTeX, PDF, or code artifacts when preview commands are visible, or fall back to shell/browser tools. Use when the user wants to review a written artifact, export a report, or view a rendered document.

Provision and manage GPU pods on RunPod for explicitly chosen long-running research experiments. Use when a Feynman replication, benchmark, or dataset-heavy research run needs persistent GPU compute with SSH access.

Use this for complex debugging, design review, deep refactors, or research-heavy coding work. The agent must inspect the repository first, plan before editing, make one coherent patch, and validate changes.

Multiagent/subagent workflow. Split work only when tasks are genuinely independent. Each worker reports files changed, tests run, and risks. Coordinator integrates and validates.

Fast mechanical mode for tiny, well-defined edits. No refactors, no abstractions, no cleanup. Return a short summary and validation command.

Default implementation mode. Make the smallest safe patch that solves the stated problem, preserve repository style, and run narrow validation.

Debugging mode for failing builds, tests, runtime errors, and bad configs. Reproduce or inspect the failure first, distinguish root cause from symptoms, prefer a minimal fix.

Expert mode for this NixOS/Home Manager repository. Prefer declarative Nix, follow local conventions, keep module boundaries clean.

Safe repository editing mode. Inspect first, touch the minimum number of files, protect user work, never commit/push/delete unless explicitly asked.

Research-heavy technical decisions. Gather evidence before recommending, prefer primary sources, separate facts from inference, produce a practical recommendation.

Code review / design review mode. Be skeptical. Identify correctness, maintainability, security, and complexity risks. Rank by severity. Propose the smallest fix for each high-impact issue.

Validation-focused mode. Discover the narrowest relevant test command, run it, report pass/fail clearly. Do not hide failures.

Look up any arxiv paper on alphaxiv.org to get a structured AI-generated overview. This is faster and more reliable than trying to read a raw PDF.