aiq

Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.

Use when asked to run deep research or AI-Q research through a reachable NVIDIA AI-Q Blueprint backend.

Use when changing AI-Q continuous integration, pre-commit, or contributor governance — editing .github/workflows/ (ci, ui, skills-eval, request-nvskills-ci), .pre-commit-config.yaml hooks, .github/CODEOWNERS, .coderabbit.yaml, copy-pr-bot, or the .github/skill-eval harness — and validating those changes without breaking the gate.

Use when validating an AI-Q change before opening or merging a PR — choosing and running the right Python, frontend, docs, or eval checks for the surfaces you touched instead of one fixed command list.

Use when adding or changing an AI-Q data source under sources/, registering it as a NeMo Agent Toolkit function, wiring it into the data_source_registry for UI toggles, or validating retrieval behavior with tests.

Use when adding or changing a general-purpose AI-Q tool (a NeMo Agent Toolkit function) under sources/, defining its FunctionBaseConfig schema, registering it with @register_function, wiring it into an agent's tools list, or testing it.

Use when composing, adapting, or validating an AI-Q workflow YAML under configs/ — selecting a shipped profile, enabling tools and data_source_registry sources, wiring agents and the chat_deepresearcher_agent workflow, configuring general.telemetry (Phoenix, LangSmith, Weave, OTEL) and general.front_end aiq_api settings, and pre-flighting cross-references before deploy or serve. Hand off deploy to aiq-deploy, live research to aiq-research, prompt/model edits to aiq-customize-prompts-models, and new source code to aiq-add-tool or aiq-add-data-source.

Use when customizing AI-Q agent behavior through Jinja2 prompt templates or per-agent model selection — editing prompts under src/aiq_agent/agents/*/prompts/, adding template variables, or assigning/swapping LLMs per agent role via config (the llms section plus per-agent fields like orchestrator_llm, planner_llm, researcher_llm, writer_llm, source_router_llm).

Use this skill to turn researched or computed numeric data into source-grounded charts (PNG) plus the underlying CSV, by writing Python/matplotlib code and running it in the job-scoped sandbox. The chart is harvested as a durable artifact and embedded in the final report. Triggers: "chart", "plot", "graph", "bar chart", "line chart", "visualize", "trend over time", "compare visually", "figure". Outputs: a PNG chart artifact, a CSV of the plotted data, and a manifest describing them.

Use this skill for converting researched facts or user-provided data into structured tables by writing code, then running Python/pandas calculations in the job-scoped sandbox. This skill is for numeric normalization, tabular analysis, rankings, growth rates, summary statistics, CSV/JSON generation, and markdown tables. Triggers: "compute table", "calculate growth", "normalize values", "extract figures", "rank companies", "QoQ", "YoY", "CAGR", "summary statistics", "CSV", "JSON", "markdown table", "standardize quarters", "standardize currencies", "compare over time". Outputs: Markdown tables, CSV text, JSON records, summary statistics, rankings, and data-quality notes.

Use this skill during research for lightweight forecast evidence analysis: base-rate checks, prediction-market anchors, scenario ranges, directional factor summaries, implied probabilities, and monitoring indicators. Triggers: "forecast", "prediction", "probability", "odds", "base rate", "scenario", "Polymarket", "prediction market", "will happen", "market-implied", "confidence interval", "price target", "expected value". Outputs: forecast evidence notes or compact forecast inputs returned in your ResearchNotes for writer synthesis.

Use this skill for small deterministic calculations during research when pandas/table analysis is unnecessary. Triggers: "calculate", "arithmetic", "unit conversion", "percentage point", "expected value", "weighted average", "range", "ratio", "sanity check", "implied value", "probability conversion". Outputs: concise calculation notes, JSON snippets, or Markdown bullets returned in your ResearchNotes for later synthesis.

Use this skill when the final answer strategy calls for a long-form report, publication-quality research writeup, comprehensive analysis, deep dive, whitepaper-style narrative, or detailed cited Markdown report. Triggers: "long_form_report", "long-form report", "comprehensive report", "publication-ready", "deep dive", "in-depth analysis", "whitepaper", "research report", "detailed report", "full report". Outputs: A polished, cited Markdown report with coherent structure, analytical depth, source-grounded claims, limitations, and a sources section.

Use when preparing, opening, or updating an AI-Q pull request — scoping the branch, signing commits with DCO, filling the PR template with real validation evidence, and following the copy-pr-bot CI and merge flow.

Use this skill when the final answer strategy calls for a prediction, forecast, probability estimate, price target, expected value, threshold outcome, scenario outlook, or prediction-style research report. This skill takes priority over general long-form writing skills for forecasts, stock-price outlooks, probabilities, odds, price targets, expected values, and next-period value estimates, even when the requested answer is concise. Triggers: "prediction", "forecast", "probability", "odds", "confidence interval", "price target", "expected value", "will X happen", "stock price", "market outlook", "scenario analysis", "threshold", "single forecast", "predict". Outputs: A cited Markdown prediction report with a clear forecast, confidence interval or range, executive summary, supporting and opposing factors, uncertainties, recommendations or implications, methodology note, and sources.