测试与安全技能

Nicht böse, nur falsch optimiert: Was OpenAIs RubyGems-Vorfall über KI-Agenten und Reward-Hacking zeigt Wie OpenAIs Agenten durch Reward-Hacking eine

中文简介正在准备中。当前显示原文说明。

PVDetector, a training-free framework that detects PI attacks during LLM inference by measuring hidd GAP: no mappable capability in latest signal (our

中文简介正在准备中。当前显示原文说明。

Prompt Injection Defense 2026: 8 Tested Techniques … — 20. Apr. 2026 · Prompt injection is still the

中文简介正在准备中。当前显示原文说明。

LLM) can be manipulated through prompt injection, whether a retrieval pipeline leaks confidential da

中文简介正在准备中。当前显示原文说明。

Prompt Injection Challenge: LLMail-Inject The competition features 40 unique levels, each combining

中文简介正在准备中。当前显示原文说明。

Use when reporting build, test, or deploy results - enforces real tool output over plausible claims.

中文简介正在准备中。当前显示原文说明。

Dan Goodin – 8/27/2026 | 157 How OpenAI let a mob of LLM agents game a test and ransack Hugging Face Without authorization, 1,200 OpenAI agents conspi

中文简介正在准备中。当前显示原文说明。

Econometrics skill for instrumental variables and treatment effect estimation. Activates when the user asks about: "instrumental variables", "IV estimation", "2SLS", "two-stage least squares", "endogeneity", "weak instruments", "first stage", "Sargan test", "overidentification", "propensity score matching", "PSM", "average treatment effect", "ATT", "LATE", "local average treatment effect", "endogenous regressor", "instrument validity", "工具变量", "两阶段最小二乘", "内生性", "弱工具变量", "倾向得分匹配", "平均处理效应", "处理效应", "局部平均处理效应"

中文简介正在准备中。当前显示原文说明。

Econometrics skill for Synthetic Control Method (SCM). Activates when the user asks about "synthetic control", "SCM", "placebo test", "synthetic DID", "合成控制", "安慰剂检验", "合成反事实", "合成DID".

中文简介正在准备中。当前显示原文说明。

Econometrics skill for Regression Discontinuity Design (RDD). Activates when the user asks about: "regression discontinuity", "RDD", "RD design", "sharp RDD", "fuzzy RDD", "running variable", "forcing variable", "cutoff", "bandwidth selection", "local linear regression", "McCrary test", "density test", "RDROBUST", "continuity assumption", "donut hole RDD", "geographic RDD", "断点回归", "回归不连续", "运行变量", "截断值", "带宽选择", "精确断点", "模糊断点", "密度检验", "局部线性回归"

中文简介正在准备中。当前显示原文说明。

Comprehensive results analysis for empirical research: generate publication-quality descriptive statistics and balance tables, interpret regression coefficients with economic magnitude and effect sizes, assess identification assumption diagnostics, and produce structured results memos. Use when asked to create summary statistics, Table 1, balance tests, interpret results, assess economic significance, or write results narratives.

中文简介正在准备中。当前显示原文说明。

Econometrics skill for time series analysis. Activates when the user asks about: "time series", "stationarity", "unit root test", "ADF test", "KPSS test", "ARIMA", "ARMA", "autocorrelation", "ACF", "PACF", "VAR model", "VECM", "Granger causality", "cointegration", "impulse response function", "forecast", "seasonal decomposition", "ARCH", "GARCH", "时间序列", "平稳性检验", "单位根", "自回归", "格兰杰因果", "协整", "脉冲响应", "预测", "向量自回归"

中文简介正在准备中。当前显示原文说明。

Analyze the current repo structure, build system, test setup, and conventions to provide a practical onboarding guide. Use when new to a codebase, joining a project, or wanting to understand how a repo is organized. Triggers on getting started, new to repo, onboard, how does this repo work, repo structure, codebase overview.

中文简介正在准备中。当前显示原文说明。

Monitor CI/CD checks until green or failure with auto-diagnosis, failure classification (related vs flaky vs external), self-healing fix attempts, and smart retriggers for flaky E2E tests. Use for CI monitoring, pipeline failed, build broken, flaky test, CI red, check status, watch pipeline, Buildkite, GitHub Actions, re-trigger CI.

中文简介正在准备中。当前显示原文说明。

Turn one approved topic into a curiosity-led 3D educational short with character sheets, ordered keyframes, motion clips, narration, captions, and an auditable assembly plan.

中文简介正在准备中。当前显示原文说明。

Knowledge base from NIST SP 800-160 Vol 1 (Engineering Trustworthy Secure Systems) and Vol 2 (Developing Cyber-Resilient Systems). Use for systems security engineering (SSE): the three SSE framework contexts (problem/solution/trustworthiness), the 30+ design principles for trustworthy secure design, trustworthiness and assurance/assurance cases, security across the ISO/IEC/IEEE 15288 life-cycle process groups, and the Vol 2 cyber-resiliency engineering framework — 4 goals (anticipate/withstand/recover/adapt), objectives, 14 techniques, implementation approaches, and design principles for the APT/adverse-conditions threat model. Bridges systems engineering and the NIST security cluster (CSF/SSDF/RMF). Does not reproduce ISO/IEC/IEEE 15288 process text, nor cover SP 800-53 control catalogs in depth.

中文简介正在准备中。当前显示原文说明。

Knowledge base from NIST Secure Software Development Framework (SSDF) v1.1 (SP 800-218). Use for secure software development practices, SDLC security integration, software supply chain security, vulnerability disclosure, SBOM, threat modelling, secure coding, build hardening, EO 14028 compliance. Covers the four SSDF practice groups (PO, PS, PW, RV) only; does not cover NIST CSF, SP 800-53 controls, or cloud-specific security frameworks in depth.

中文简介正在准备中。当前显示原文说明。

Knowledge base from the NIST Framework for Cyber-Physical Systems (SP 1500-201/202/203, v1.0, 2017). Use for engineering and analyzing cyber-physical systems (CPS) and IoT: the CPS Framework's aspect × facet grid (nine aspects of concern examined through conceptualization, realization, and assurance facets), cross-property trustworthiness risk management (safety/security/privacy/reliability/resilience together), data interoperability (syntactical/semantic/contextual, canonical models, identifiers, provenance), the timing aspect (time-interval classes, time domains, time-aware networking) and secure/resilient time (GNSS jamming/spoofing, PTP/NTP attacks, holdover, the clock-error model and Allan/MTIE metrics), plus the use-case requirements method. Bridges IT and OT under a clock. Scope limits: a 2017 living/draft document, thin on post-2017 standards and on detailed control catalogs; it names but does not reproduce ISO/IEC/IEEE 42010 and 15288 process text or permission-licensed third-party figures.

中文简介正在准备中。当前显示原文说明。

Knowledge base from NIST SP 800-61 Rev. 3 (Incident Response Recommendations for Cybersecurity Risk Management). Use for CSF 2.0-aligned incident response life cycles, IR roles, Community Profile priorities for DE/RS/RC and preparation, playbooks, coordination, and continuous improvement. Covers 800-61r3 only; does not replace CSF 2.0 full text, SP 800-61r2 tactical playbooks, or sector-specific breach-notification law.

中文简介正在准备中。当前显示原文说明。

Synthesized reference notes from three US Government vehicle-safety sources: NHTSA Cybersecurity Best Practices for the Safety of Modern Vehicles (updated 2022, final), NHTSA Automated Driving Systems 2.0: A Vision for Safety (2017), and a 16-section selection of 49 CFR Part 571 (FMVSS) pinned at eCFR versioner date 2025-01-01. Use for voluntary vehicle cybersecurity practices, the ADS 2.0 twelve priority safety design elements, and FMVSS occupant protection, brakes/stability/lighting, and fuel/EV integrity requirements. Not ISO 26262, not ASPICE, not ISO/SAE 21434, not UNECE R155/R156, not SAE J3016, and not the full Part 571; the FMVSS material is a selection of vehicle-level minimum performance standards, not a process standard.

中文简介正在准备中。当前显示原文说明。

Knowledge base from NIST SP 800-37 Revision 2, the Risk Management Framework (RMF) for information systems and organizations. Use for the seven RMF steps (Prepare, Categorize, Select, Implement, Assess, Authorize, Monitor) and their tasks; the three-tier organization-wide risk model; authorization boundaries; the requirements-to-controls relationship; security categorization (FIPS 199/200 high-water mark) and control selection/tailoring (SP 800-53/53B baselines); assessment, the authorization decision and risk acceptance, continuous monitoring and ongoing authorization; the security/privacy distinction; and supply chain risk management. This is the security-and-privacy-risk-governance edge of systems engineering — it governs how much protection and whether to accept residual risk, complementing the disciplines that decide how protection is built. Does NOT reproduce the SP 800-53 control catalog or its baselines, the FIPS impact tables, or the ISO/IEC/IEEE 15288 process text; it names and routes to them rather than restating them.

中文简介正在准备中。当前显示原文说明。

Knowledge base from NIST Cybersecurity Framework (CSF) 2.0. Use for cybersecurity risk management, GOVERN, IDENTIFY, PROTECT, DETECT, RESPOND, RECOVER, Organizational Profiles, CSF Tiers, supply chain risk, enterprise risk management, incident response. Covers the CSF 2.0 document only; does not cover specific SP 800-53 controls, NIST RMF system authorization steps, or sector-specific regulations.

中文简介正在准备中。当前显示原文说明。

Knowledge base from the NIST/SEMATECH e-Handbook of Statistical Methods (NIST HB 151) — the practical statistics reference for engineering, metrology, and quality. Use for: exploratory data analysis and the four univariate assumptions (4-plot); measurement process characterization including bias/precision, calibration designs, gauge R&R, and ISO/GUM uncertainty budgets; production process characterization (stability vs. capability); process modeling and regression (LS/WLS/NLS/LOESS); design of experiments (screening, fractional/factorial, response-surface, Taguchi); statistical process control (Shewhart/CUSUM/EWMA charts, capability indices, acceptance sampling); product/process comparisons (hypothesis tests and confidence intervals for 1/2/3+ groups, ANOVA, multiple comparisons); and reliability (lifetime & repair-rate models, accelerated testing, reliability growth). Scope limits: this is applied frequentist statistics for measurement and quality — it does NOT reproduce the per-distribution formula galleries, worked case studies, datasets, plot images, or Dataplot/R code of the original web Handbook (those are described, not copied); it is thin on modern machine learning, Bayesian methods beyond conjugate reliability priors, time-series/forecasting, and Bayesian experimental design.

中文简介正在准备中。当前显示原文说明。

Knowledge base from the GAO Cost Estimating and Assessment Guide (GAO-20-195G). Use for program cost estimating: the four characteristics of a reliable estimate (comprehensive, well documented, accurate, credible) and the 18 best practices; the 12-step cost estimating process (purpose, plan, technical baseline, WBS, ground rules and assumptions, data, point estimate, sensitivity, risk/uncertainty, document, present, update); estimating methods (analogy, parametric, engineering build-up, learning curves); Monte Carlo risk/uncertainty analysis, confidence levels and contingency; auditing and validating an estimate against the characteristics; earned value management (EIA-748, BCWS/BCWP/ACWP, CPI/SPI/TCPI, EAC, PMB, IBR); and specialized techniques (software cost estimating, learning curves, Analysis of Alternatives, WBS templates, the Green Book internal-control framework). This is the cost-estimating companion to the GAO Schedule and Technology Readiness Assessment guides. Does not cover schedule estimating/risk in depth (see the GAO Schedule Assessment Guide), agency-specific cost models, or the full text of referenced standards (EIA-748, MIL-STD-881D, the Green Book) — these are named, not reproduced.

中文简介正在准备中。当前显示原文说明。