Research

- RL with policy advice. Azar et al., ECML 2013. - Reduction from RL to bandit problem. - Regret bounds: sum of differences between actual policy and optimal policy. - Regret scales with the number of tasks \sqrt(M), rather than the state and action space. - Brunskill and Li, UAI 2013. Reduce from RL to (active) classification problem. - https://cs.stanford.edu/people/ebrun - Provably speeding multitask RL. Guo and Brunskill, AAAI 2015. K tasks sampled from M tasks. Evaluation goal: provably ...

발굴조사 보고서 고찰(考察) 섹션 초안 작성

Compare FinOps metrics across multiple repositories in an organization

Audit Claude subagent configs for completeness, security, and best practices. Use when reviewing agents/ for missing frontmatter, overprivileged tools, or bad model choices.

Design and launch a product survey and produce a Survey Pack (brief, questionnaire/instrument, analysis plan, launch checklist, reporting outline). Use for customer surveys, onboarding surveys, NPS/CSAT/PMF, cancellation/churn, and feedback surveys.

Deep multi-source research combining academic MCPs (arxiv, semantic-scholar, paper-search, deepwiki), Exa semantic search, and local ~/.topos knowledge base. Use for comprehensive research requiring multiple sources. NEVER fall back to web_search - ask user for help instead.

提供v3_e-report数据库的SQL查询模板，包括数据分析告警配置等表的查询。查询分析告警配置时使用。使用 exec_sql 工具执行查询。

Datalog bottom-up fixpoint iteration for recursive queries

Expert data researcher specializing in discovering, collecting, and analyzing diverse data sources. Masters data mining, statistical analysis, and pattern recognition with focus on extracting meaningful insights from complex datasets to support evidence-based decisions.

Systematic data validation, error detection, cross-source reconciliation, and query correctness checking for analytical work. Use when validating Snowflake queries, catching calculation errors, reconciling metrics across different data sources, checking for null values, ensuring date range validity, detecting statistical anomalies, validating metric calculations (median vs mean, rate normalization), checking aggregation grain (per-record vs per-entity), validating contribution analysis for no...

Use when extracting structured data from medical research PDFs, parsing study characteristics, patient demographics, outcomes, and results. Invoke for systematic review data collection from papers.

Explore and analyze pilot data sets to uncover patterns, anomalies, and initial insights. Use when performing ad-hoc data investigations, validating data quality, or preparing exploratory visualizations for hypothesis generation.

Inventory available datasets, instrumentation gaps, and data quality considerations for the initiative.

Advanced data analysis, pattern detection, and insight generation from structured and unstructured datasets

Analyze CS2 demo files and generate comprehensive tactical reports

Analysis of crypto-to-crypto vs crypto-to-fiat (USD) trading on Alpaca for fee optimization and risk considerations

Simplify crypto hard filters to essential checks only. Trigger when: (1) crypto symbols fail multiple filters, (2) data_quality/spread/trading_status fail for crypto, (3) yfinance data gaps causing false failures.

Pattern for aggregating insights across multiple tasks to enable data-driven evolution.

Chemical Reaction Network topology for generating and analyzing reaction

This skill should be used when validating user prompts, plans, or requests before execution. It conducts independent validation, questions assumptions, references official sources and documentation, identifies potential issues, and presents revised recommendations to ensure accuracy and completeness.

Engage in critical thinking by questioning assumptions, exploring alternative perspectives, and uncovering latent topics in conversations. Use when discussions could benefit from deeper exploration, when identifying blind spots, or when broadening understanding through respectful challenge and curiosity-driven inquiry.

This skill enables honest, pressure-tested feedback on ideas, decisions, and proposals. Use this skill when prompted for an opinion on whether something is a good idea, should be done, or what to think about an approach.

Generate constructive criticism reviews for popular tech articles. Use when asked to review, critique, or analyze trending tech articles from platforms like Qiita, Zenn, Hacker News, or Dev.to. (user)

Applies rigorous critical analysis to evaluate claims, arguments, and research. Use when evaluating evidence quality, peer reviewing content, assessing argument validity, or identifying weaknesses in reasoning. Triggers on phrases like "critically analyze", "evaluate this", "review this paper", "check the logic", "assess the evidence", "find flaws", "peer review".