AI & ML

pjt222 Skill

Fit hidden Markov models using the Baum-Welch (EM) algorithm with model selection, Viterbi decoding for state sequences, and forward-backward probabilities. Use when observations are generated by unobservable latent states, you need to segment a time series into latent regimes (market regimes, speech phonemes, biological sequences), compute sequence probabilities, decode the most likely hidden state path, or compare models with different numbers of hidden states.

pjt222 Skill

Install skills, agents, and teams from agent-almanac into any supported agentic framework using the CLI. Covers framework detection, content search, installation with dependency resolution, health auditing, and manifest-based syncing. Use when setting up a new project with agentic capabilities, installing specific skills or entire domains, targeting multiple frameworks simultaneously, or maintaining a declarative manifest of installed content.

pjt222 Skill

Create a Dockerfile for R projects using rocker base images. Covers system dependency installation, R package installation, renv integration, and optimized layer ordering for fast rebuilds. Use when containerizing an R application or analysis, creating reproducible R environments, deploying R-based services (Shiny, Plumber, MCP server), or setting up consistent development environments across machines.

pjt222 Skill

Set up systematic data labeling workflows using Label Studio or similar tools. Implement quality controls, measure inter-annotator agreement, manage labeler teams, and integrate labeled data into ML training pipelines. Use when starting a supervised ML project that requires labeled training data, when model performance is limited by insufficient labeled examples, when labeling text, images, audio, or video, or when implementing active learning to prioritize the most valuable examples.

pjt222 Skill

Monitor, cap, and recover from context accumulation in agentic systems. Covers per-cycle cost tracking, context window auditing, budget caps with enforcement policies, emergency pruning when approaching limits, and progressive disclosure integration to minimize token spend on routing. Use when running long-lived agent loops (heartbeats, polling, autonomous workflows), when context windows are growing unpredictably between cycles, when API costs spike beyond expected baselines, when designing new agentic workflows that need cost guardrails from the start, or when post-mortem analysis reveals a cost incident caused by context accumulation.

pjt222 Skill

Implement comprehensive model drift monitoring using Evidently AI, statistical tests (PSI, KS), and custom metrics to detect data drift and concept drift in production ML systems. Set up automated alerting and reporting workflows to catch degradation before it impacts business metrics. Use when production models show unexplained performance degradation, when new data distributions differ from training data, when seasonal shifts affect input features, or when regulatory requirements mandate model monitoring.

pjt222 Skill

Design monochrome ornamental patterns grounded in Alexander Speltz's classical ornament taxonomy. Covers historical period selection, motif structural analysis, prompt construction for line art and silhouette rendering, and AI-assisted image generation via Z-Image. Use when creating decorative borders, medallions, or friezes in a single color, exploring historical ornament styles through generative AI, producing line art or pen-and-ink renderings of classical motifs, or generating reference imagery for design or educational materials.

pjt222 Skill

Export and optimize 3D models for FDM/SLA printing including STL/3MF export, mesh integrity verification, wall thickness checking, support generation, and slicing. Use when exporting from CAD or modeling software for 3D printing, verifying STL/3MF files are printable before slicing, troubleshooting models that fail to slice correctly, optimizing part orientation for strength or surface finish, or converting between model formats while preserving printability.

pjt222 Skill

Scaffold a new MCP server from tool specifications using the official SDK (TypeScript or Python), including transport configuration, tool handlers, and test harness. Use when you have a tool specification and need a working server, when starting a new MCP server project and want correct structure from the start, when migrating an existing tool integration to the MCP protocol, or when prototyping a tool surface to test with Claude Code before full implementation.

pjt222 Skill

Validate deliverables and build evidence trails when work passes between agents. Covers expected outcome specification before execution, structured evidence generation during execution, deliverable validation against external anchors after execution, fidelity checks for compressed or summarized outputs, trust boundary classification, and structured disagreement reporting on verification failure. Use when coordinating multi-agent workflows, reviewing cross-agent handoffs, producing external-facing outputs, or auditing whether an agent's summary faithfully represents its source material.

pjt222 Skill

Build and analyze discrete or continuous Markov chains including transition matrix construction, state classification, stationary distribution computation, and mean first passage times. Use when modeling a memoryless system with observed transition counts or rates, computing long-run steady-state probabilities, determining expected hitting times or absorption probabilities, classifying states as transient or recurrent, or building a foundation for hidden Markov models or reinforcement learning MDPs.

pjt222 Skill

Audit, classify, and selectively forget stored memories. Covers memory enumeration and classification by type/age/access frequency, staleness detection for outdated references, fidelity checks using external anchors, a decision tree for selective deletion, counter-memory inoculation for failed strategies that would otherwise be re-derived, preemptive filtering rules for what should never become memories, and an audit trail so forgetting itself is reviewable. Use when memory has grown large and uncurated, when project state has shifted significantly since memories were written, when retrieval quality has degraded, or as periodic maintenance alongside manage-memory.

pjt222 Skill

Design and execute A/B tests for ML models in production using traffic splitting, statistical significance testing, and canary/shadow deployment strategies. Measure performance differences and make data-driven decisions about model rollout. Use when validating a new model version before full rollout, comparing candidate models trained with different algorithms, measuring business metric impact of model changes, or when regulatory requirements mandate gradual rollout.

pjt222 Skill

Launch all available agents in parallel waves for open-ended hypothesis generation on problems where the correct domain is unknown. Use when facing a cross-domain problem with no clear starting point, when single-agent approaches have stalled, or when diverse perspectives are more valuable than deep expertise. Produces a ranked hypothesis set with convergence analysis and adversarial refinement.

pjt222 Skill

Analyze an arbitrary codebase to identify functions, APIs, and data sources suitable for exposure as MCP tools, producing a tool specification document. Use when planning an MCP server for an existing project, auditing a codebase before wrapping it as an AI-accessible tool surface, comparing what a codebase can do versus what is already exposed via MCP, or generating a tool spec to hand off to scaffold-mcp-server.

pjt222 Skill

Consistent agent behavior after restart — progressive identity loading, working context reconstruction from persistent artifacts, fresh-vs-continuation detection, calibration through centering and attunement, and identity verification for coherence. Addresses the cold-start problem where an agent must reconstruct who it is and what it was doing from evidence rather than memory. Use at the start of every new session, after a session interruption or crash, when agent behavior feels inconsistent with prior sessions, or when persistent memory and current context appear contradictory.

pjt222 Skill

Assess an open-source agent framework for investment readiness by evaluating community health, supersession risk, architecture alignment, and governance sustainability. Produces a four-tier classification (INVEST / EVALUATE-FURTHER / CONTRIBUTE-CAUTIOUSLY / AVOID) to guide resource allocation decisions before committing engineering effort.

pjt222 Skill

Fit cognitive drift-diffusion models (Ratcliff DDM) to reaction time and accuracy data with parameter estimation (drift rate, boundary separation, non-decision time), model comparison, and parameter recovery validation. Use when modeling binary decision-making with reaction time data, estimating cognitive parameters from experimental data, comparing sequential sampling model variants, or decomposing speed-accuracy tradeoff effects into latent cognitive components.

pjt222 Skill

Register trained models in MLflow Model Registry with version control, implement stage transitions (Staging, Production, Archived) with approval workflows, and manage model lineage with comprehensive metadata and deployment tracking. Use when promoting a trained model from experimentation to production, managing multiple model versions across development stages, implementing approval workflows for governance, rolling back to previous versions, or auditing model changes for compliance.

pjt222 Skill

Translate agent-almanac content (skills, agents, teams, guides) into a target locale while preserving code blocks, IDs, and technical structure. Covers scaffolding, frontmatter setup, prose translation, code preservation, and freshness tracking. Use when localizing content for a new language, updating stale translations after source changes, or batch-translating a domain.

pjt222 Skill

Apply collective intelligence coordination patterns — stigmergy, local rules, and quorum sensing — to organize distributed systems, teams, or workflows without centralized control. Covers signal design, agent autonomy boundaries, emergent behavior cultivation, and feedback loop tuning. Use when designing distributed systems without a coordination bottleneck, organizing teams that must self-coordinate, building event-driven architectures with shared state communication, or replacing fragile centralized orchestration with resilient emergent coordination.

pjt222 Skill

Build a custom MCP (Model Context Protocol) server that exposes domain-specific tools to AI assistants. Covers server implementation in Node.js or R, tool definitions, transport configuration, and testing with Claude Code. Use when you need to expose custom functionality beyond what mcptools provides, when building specialized domain-specific AI integrations, or when wrapping existing APIs or services as MCP tools.

pjt222 Skill

使用 Baum-Welch（EM）算法拟合隐马尔可夫模型，包含模型选择、用于状态序列的 Viterbi 解码以及前向-后向概率。适用于观测由不可观测的隐状态生成时、需要将 时间序列分割为隐含状态（市场状态、语音音素、生物序列）时、计算序列概率、 解码最可能的隐状态路径，或比较不同隐状态数目的模型。

pjt222 Skill

使用 CLI 将 agent-almanac 中的技能、代理和团队安装到任何受支持的代理 框架中。涵盖框架检测、内容搜索、带依赖解析的安装、健康审计和基于 manifest 的同步。在设置带代理能力的新项目、安装特定技能或整个领域、 同时针对多个框架，或维护已安装内容的声明性 manifest 时使用。

pjt222 Skill

Register trained models in MLflow Model Registry with version control, implement stage transitions (Staging, Production, Archived) with approval workflows, and manage model lineage with comprehensive metadata and deployment tracking. Use when promoting a trained model from experimentation to production, managing multiple model versions across development stages, implementing approval workflows for governance, rolling back to previous versions, or auditing model changes for compliance.

pjt222 Skill

Translate agent-almanac content (skills, agents, teams, guides) into a target locale while preserving code blocks, IDs, and technical structure. Covers scaffolding, frontmatter setup, prose translation, code preservation, and freshness tracking. Use when localizing content for a new language, updating stale translations after source changes, or batch-translating a domain.

pjt222 Skill

Apply collective intelligence coordination patterns — stigmergy, local rules, and quorum sensing — to organize distributed systems, teams, or workflows without centralized control. Covers signal design, agent autonomy boundaries, emergent behavior cultivation, and feedback loop tuning. Use when designing distributed systems without a coordination bottleneck, organizing teams that must self-coordinate, building event-driven architectures with shared state communication, or replacing fragile centralized orchestration with resilient emergent coordination.

pjt222 Skill

Implement a JSON-RPC 2.0 A2A server with full task lifecycle management (submitted/working/completed/failed/canceled/input-required), SSE streaming, and push notifications. Use when implementing an agent that participates in multi-agent A2A workflows, building a backend for an Agent Card, adding A2A protocol support to an existing agent or service, or deploying an agent that must interoperate with other A2A-compliant agents.

pjt222 Skill

Create a Dockerfile for R projects using rocker base images. Covers system dependency installation, R package installation, renv integration, and optimized layer ordering for fast rebuilds. Use when containerizing an R application or analysis, creating reproducible R environments, deploying R-based services (Shiny, Plumber, MCP server), or setting up consistent development environments across machines.

pjt222 Skill

Set up systematic data labeling workflows using Label Studio or similar tools. Implement quality controls, measure inter-annotator agreement, manage labeler teams, and integrate labeled data into ML training pipelines. Use when starting a supervised ML project that requires labeled training data, when model performance is limited by insufficient labeled examples, when labeling text, images, audio, or video, or when implementing active learning to prioritize the most valuable examples.

pjt222 Skill

Monitor, cap, and recover from context accumulation in agentic systems. Covers per-cycle cost tracking, context window auditing, budget caps with enforcement policies, emergency pruning when approaching limits, and progressive disclosure integration to minimize token spend on routing. Use when running long-lived agent loops (heartbeats, polling, autonomous workflows), when context windows are growing unpredictably between cycles, when API costs spike beyond expected baselines, when designing new agentic workflows that need cost guardrails from the start, or when post-mortem analysis reveals a cost incident caused by context accumulation.

pjt222 Skill

Implement comprehensive model drift monitoring using Evidently AI, statistical tests (PSI, KS), and custom metrics to detect data drift and concept drift in production ML systems. Set up automated alerting and reporting workflows to catch degradation before it impacts business metrics. Use when production models show unexplained performance degradation, when new data distributions differ from training data, when seasonal shifts affect input features, or when regulatory requirements mandate model monitoring.

pjt222 Skill

Design monochrome ornamental patterns grounded in Alexander Speltz's classical ornament taxonomy. Covers historical period selection, motif structural analysis, prompt construction for line art and silhouette rendering, and AI-assisted image generation via Z-Image. Use when creating decorative borders, medallions, or friezes in a single color, exploring historical ornament styles through generative AI, producing line art or pen-and-ink renderings of classical motifs, or generating reference imagery for design or educational materials.

pjt222 Skill

Export and optimize 3D models for FDM/SLA printing including STL/3MF export, mesh integrity verification, wall thickness checking, support generation, and slicing. Use when exporting from CAD or modeling software for 3D printing, verifying STL/3MF files are printable before slicing, troubleshooting models that fail to slice correctly, optimizing part orientation for strength or surface finish, or converting between model formats while preserving printability.

pjt222 Skill

Scaffold a new MCP server from tool specifications using the official SDK (TypeScript or Python), including transport configuration, tool handlers, and test harness. Use when you have a tool specification and need a working server, when starting a new MCP server project and want correct structure from the start, when migrating an existing tool integration to the MCP protocol, or when prototyping a tool surface to test with Claude Code before full implementation.

pjt222 Skill

Validate deliverables and build evidence trails when work passes between agents. Covers expected outcome specification before execution, structured evidence generation during execution, deliverable validation against external anchors after execution, fidelity checks for compressed or summarized outputs, trust boundary classification, and structured disagreement reporting on verification failure. Use when coordinating multi-agent workflows, reviewing cross-agent handoffs, producing external-facing outputs, or auditing whether an agent's summary faithfully represents its source material.