Latest Agent Skills

Track ML experiments, manage model registry with versioning, deploy models to production, and reproduce experiments with MLflow.

Build production RAG and semantic search systems with a high-performance vector database written in Rust, supporting hybrid search, filtering, and horizontal scaling.

Store and query embeddings with metadata filtering, vector search, and full-text search using an open-source database that scales from notebooks to production.

Enables fast similarity search and clustering of dense vectors using FAISS, supporting billions of vectors, GPU acceleration, and various index types.

Improves multi-step research reliability with structured protocols for source validation, error recovery, and transparent reasoning.

Debug and optimize AI agents by analyzing reasoning traces, context degradation, tool confusion, instruction drift, repeated task failures, and performance regressions.

Manages a personal operating system for content creation, relationship management, goal tracking, and weekly reviews using progressive disclosure and append-only data files.

Design multi-agent systems with context isolation, supervisor or swarm coordination, explicit handoffs, parallel execution, and decision frameworks for when multiple agents are justified.

Extends effective context capacity through strategic compression, masking, caching, and partitioning techniques.

Explains foundational concepts of context engineering: what context is, attention mechanics, the U-shaped attention curve, and why context quality matters more than quantity.

Guides project-level decisions for LLM-powered systems: task-model fit, pipeline architecture, token and cost estimation, and agent-assisted iteration.

Designs autonomous agent harnesses with locked evaluators, editable surfaces, durable logging, novelty gates, pruning, rollback, and human approval boundaries.

Diagnose and mitigate context degradation patterns including lost-in-middle failures, context poisoning, distraction, confusion, and clash in AI agent systems.

Optimizes long-running agent sessions with structured context compression, summarization, and durable handoff summaries that preserve decisions, files, risks, and next actions.

Provides production-grade techniques for evaluating LLM outputs using LLMs as judges, covering direct scoring, pairwise comparison, bias mitigation, rubric generation, and confidence calibration.

Convert books into supervised fine-tuning datasets and train style-transfer models that replicate an author's voice.

Uses the filesystem as a durable overflow layer for agent context, enabling scratch pads, plan persistence, sub-agent communication, dynamic skill loading, and log management.

Model agent mental states using BDI (Beliefs, Desires, Intentions) ontology patterns, enabling cognitive reasoning, explainability, and semantic interoperability in multi-agent systems.

Shares memory between agents at the representation level by compacting the orchestrator's KV cache for efficient worker handoff, reducing token costs without summarization or retrieval.

Designs persistent memory architectures for AI agents, covering cross-session knowledge retention, entity tracking, temporal validity, graph/vector retrieval, and memory consolidation.

Design and build hosted agent infrastructure with sandboxed execution, warm pools, session persistence, and multiplayer collaboration.

Design tool descriptions, schemas, and response formats that agents can reliably route on, including MCP server design, tool-set consolidation, and naming conventions.

Build evaluation frameworks for agent systems with deterministic checks, regression suites, multi-dimensional rubrics, quality gates, production monitoring, and outcome measurement.

Provides a structured template for creating new Agent Skills, including sections for description, activation criteria, core concepts, practical guidance, and integration notes.

Provides structured guidance for building production-grade AI agent systems through context engineering, covering fundamentals, architectural patterns, operational excellence, and evaluation.

Evaluate SaaS unit economics and capital efficiency to determine if a business can scale efficiently or needs correction.

Run a customer journey mapping workshop with adaptive questions to map stages, actions, emotions, pain points, and opportunities for a persona and scenario.

Plan tailored engagement for a specific stakeholder before critical conversations, milestones, or relationship repairs.

Structure spoken product-sense interview answers with assumptions, segmentation, pain-point prioritization, and MVP tradeoffs.

Evaluate the financial impact of pricing changes using ARPU, conversion, churn risk, NRR, and payback analysis to make data-driven go/no-go decisions.

Structure the first 90 days of a VP or CPO transition as a diagnostic process, not an execution sprint, using a three-phase playbook to build evidence before making decisions.

Run a user story mapping workshop with adaptive questions and a structured map output, generating backbone activities, user tasks, and release slices for a workflow.

Calculate SaaS revenue, retention, and growth metrics to diagnose momentum, churn, expansion, or product-market-fit signals.

Diagnose whether an AI workflow suffers from context stuffing or benefits from context engineering, and apply structured techniques to improve reliability.

Assess whether your product work is AI-first or AI-shaped, evaluate maturity across five key competencies, and get recommendations on which capability to build next.

Design multi-agent AI workflows with clear boundaries, handoffs, and monitoring for complex product management tasks.

Evaluate acquisition channels using unit economics, customer quality, and scalability to decide whether to scale, test, or kill a growth channel.

Map every stakeholder before engaging anyone, using an equity-aware process that captures not just obvious sponsors but gatekeepers, impacted communities, and overlooked voices.

Evaluate feature investments using revenue impact, cost structure, ROI, and strategic value to make data-driven build/don't-build decisions.

Guide PMs through the transition to Director with adaptive coaching across preparing, interviewing, landing, and recalibrating phases.

Identify what to measure, test, or track to de-risk a product or AI idea across internal and external risk dimensions.

Diagnose SaaS business health across growth, retention, unit economics, and capital efficiency. Use when preparing a business review or prioritizing urgent fixes.

Diagnose the PM-to-Director transition using altitude (scope) and horizon (time) axes, identify transition zones, and apply the Cascading Context Map when organizational direction is vague.

Build an Opportunity Solution Tree from outcomes to opportunities, solutions, and tests. Use when a stakeholder request needs problem framing before you decide what to build.

Guides Directors and senior product leaders through the transition to VP or CPO with adaptive coaching across preparing, interviewing, landing, and recalibrating.

Turn rough notes, workshop content, or prompt dumps into compliant, publish-ready SKILL.md files that pass validation and integrate with repo documentation.

Guide product managers through a comprehensive strategy session covering positioning, discovery, and roadmap planning to move from vague direction to validated strategy.

Look up SaaS finance metrics, formulas, and benchmarks fast. Use when you need a quick metric definition, formula, or benchmark during analysis.

Plan customer discovery interviews with the right goal, segment, constraints, and method for problem validation, churn research, or new product ideas.

Calculate TAM, SAM, and SOM with explicit assumptions, methods, and caveats for market sizing.

Guide teams through MITRE's Problem Framing Canvas to clarify problem statements before jumping to solutions.

Choose a prioritization framework based on product stage, team context, and stakeholder needs.

Diagnose which organic growth path to pursue by identifying where the growth constraint lives, using a fast triage based on the McKinsey Growth Pyramid.

Guide product managers through breaking down epics into user stories using the Humanizing Work methodology with 9 splitting patterns.

Provides a canonical facilitation protocol for interactive skills: one step at a time with progress tracking, adaptive recommendations, and interruption handling.

Evaluate an AI product idea across outcomes, hypotheses, risks, and positioning to decide whether it deserves investment or recommendation.

Create a Geoffrey Moore-style positioning statement to clarify target customers, unmet needs, product category, and differentiation from alternatives.

Break large user stories or epics into smaller, independently deliverable stories using systematic splitting patterns.

Guide product managers through discovering and articulating product positioning by asking adaptive questions about target customers, unmet needs, product category, benefits, and competitive differentiation.

Create a customer journey map across stages, touchpoints, actions, emotions, and metrics to diagnose broken experiences and align teams on the full customer flow.