数据与 AI技能

Analyze rejected plans and reviewer feedback to identify recurring concerns, changing expectations, and concrete prompt improvements. Produce an HTML dashboard from Plannotator archives, or use Claude Code plan rejection reasons when those archives are unavailable.

中文简介正在准备中。当前显示英语简介。

• Finding recurring reasons for plan rejection
• Understanding how review expectations changed
• Deriving planning instructions from past feedback

Build evaluation suites for AI features with the Axiom AI SDK. Generate test files, scoring functions, test cases, and configurable parameters from descriptions and existing code to compare changes and monitor output quality in production.

中文简介正在准备中。当前显示英语简介。

• Add regression evaluations for an AI feature
• Score classification, retrieval, and structured outputs
• Compare model and parameter configurations

Translates Splunk SPL queries into Axiom APL, the query language used to search and analyze data in Axiom. Covers command and function equivalents, time filters, type conversions, and common patterns for migration or learning.

中文简介正在准备中。当前显示英语简介。

• Migrating Splunk queries to Axiom
• Converting log aggregation and error-rate queries
• Learning APL through familiar SPL patterns

Add an AI agent backend to a Convex app with persistent conversations, streamed messages, and tool calls. It also supports document retrieval for grounded answers, using Convex AI Gateway for model calls where available.

中文简介正在准备中。当前显示英语简介。

• Add an AI conversation feature to a Convex app
• Store conversation threads and message history
• Build answers using retrieved documents

Turn supplied CSV, Excel, or JSON data into a visual report page with key metric cards, multiple charts, and a data table. Include the reporting period, data sources, written observations, and a collapsible explanation of the methods used.

中文简介正在准备中。当前显示英语简介。

• Turning weekly performance data into a report page
• Reviewing key metrics and year-over-year changes
• Sharing data trends with charts and supporting tables

Turn A/B tests and product experiment records into decision reports. Organize hypotheses, setup, and metrics, explain findings and uncertainty, then recommend shipping, iterating, extending, or stopping, with follow-up experiments based only on supplied data.

中文简介正在准备中。当前显示英语简介。

• Deciding whether to launch a tested product change
• Reviewing growth or pricing experiments
• Assessing onboarding changes or email experiments

Organize parsed documents into four color-coded cards covering facts, explanations, risks, and actions. Preserve source references and generation details so readers can review evidence, identify uncertainty, and consider next steps in research, financial, or legal analysis.

中文简介正在准备中。当前显示英语简介。

• Separating facts from interpretations in financial reports
• Organizing evidence and risks in legal documents
• Turning research documents into evidence-linked analysis cards

Turn data findings into reports and presentations that explain why the numbers matter. The skill combines context, problems, recommendations, and visual guidance to help executives and nontechnical audiences understand evidence and decide on next steps.

中文简介正在准备中。当前显示英语简介。

• Explaining analysis results to executives
• Structuring quarterly business reviews
• Building investor presentation narratives

Guide an AI agent through ClawWork, an economic survival simulation where task rewards must cover API token costs. Defines a daily choice between work and learning, artifact submission, knowledge storage, and priorities based on the remaining balance.

中文简介正在准备中。当前显示英语简介。

• Choose work or learning based on the remaining balance
• Create and submit task artifacts for evaluation
• Save research findings to persistent memory

Classify messages, rate content, or estimate probabilities using defined question types. Receive machine-readable answers, with confidence scores for classification and ratings, and apply the same questions to multiple inputs in a batch.

中文简介正在准备中。当前显示英语简介。

• Routing customer inquiries to departments
• Rating support quality against defined criteria
• Classifying sentiment across multiple messages

Review how an AI agent loads, retains, summarizes, and shares working information. This historical guide helps examine forgotten constraints, repeated actions, tool selection, and handoffs using context engineering practices published in 2024–2025.

中文简介正在准备中。当前显示英语简介。

• Reviewing information loaded at startup versus on demand
• Investigating repeated actions or forgotten constraints
• Planning summaries and persistent task notes

Helps an agent take over an existing role after a planned handover. It guides checks of inherited records, identity, authority, and recurring duties, while explaining how to recover the predecessor’s reasoning and preserve lessons for the next agent.

中文简介正在准备中。当前显示英语简介。

• Getting oriented after a planned agent handover
• Checking inherited records and recurring duties
• Recovering the reasoning behind earlier decisions

Guides the design and review of agent startup context, including role files and links to authoritative records. It helps identify stale guidance, missing updates, overloaded startup files, and instructions that lose the purpose of the work.

中文简介正在准备中。当前显示英语简介。

• Designing startup files for a new agent
• Reviewing stale guidance used by running agents
• Separating role information from reusable procedures

Diagnose why an AI agent skips tools, passes incorrect arguments, or repeats calls. Get guidance on tool descriptions, schemas that define accepted inputs, execution flow, and mock tests before connecting to a real model.

中文简介正在准备中。当前显示英语简介。

• Finding why an agent never calls a tool
• Fixing incorrect or missing tool arguments
• Investigating repeated calls or skipped steps

Connect stock data to an application using Finnhub Stock API. The skill covers current quotes, company profiles, historical prices, and market news, with server-side request handling to protect API keys and manage cached responses.

中文简介正在准备中。当前显示英语简介。

• Display current stock prices on a dashboard
• Retrieve historical prices and trading volumes
• Show company profiles alongside stock quotes

Continue analysis after repeated tool failures by embedding known data in Python scripts run through run_shell. Save intermediate results as JSON with timestamps and source metadata, creating records that can be checked and reused in later steps.

中文简介正在准备中。当前显示英语简介。

• Continuing analysis when search and execution tools fail
• Comparing prices using competitor data already available
• Saving intermediate results for later review

Draft copy-ready /goal commands for long-running tasks in Claude Code and Codex. Turn a request into a checkable outcome with verification steps and stopping conditions, while flagging missing details instead of inventing them.

中文简介正在准备中。当前显示英语简介。

• Prepare a goal for a long-running development task
• Clarify completion criteria for an existing goal
• Shorten a goal prompt to fit the character limit

Diagnose why an AI workflow becomes inconsistent despite receiving extensive information. Through guided questions, define what to retain, retrieve when needed, or exclude, and develop a memory structure and a research-to-implementation process with focused context.

中文简介正在准备中。当前显示英语简介。

• Diagnosing vague answers despite extensive input
• Defining information boundaries between agents
• Planning persistent memory and on-demand retrieval

Plan NVIDIA NuRec workflows that turn sensor recordings into renderable 3D scenes. This routing skill selects the relevant specialist skills for dataset downloads, conversion, training, rendering, and frame cleanup, and puts the steps in order.

中文简介正在准备中。当前显示英语简介。

• Plan a reconstruction workflow from sensor recordings
• Find the steps to render an existing USDZ scene
• Choose a workflow for cleaning up rendered frames

Analyze product usage in PostHog and manage experiments, dashboards, surveys, and feature flags, which control feature availability. Use connected tools to investigate user behavior and errors, run queries, and summarize findings with numbers and links.

中文简介正在准备中。当前显示英语简介。

• Analyze conversion funnels and user retention
• Manage feature rollouts and A/B tests
• Investigate errors and review session recordings

Coordinate CAD and source assets through conversion to USD, a 3D scene format, material and physics assignment, and SimReady checks. Produce stage reports, with preview rendering and validated delivery packages when requested.

中文简介正在准备中。当前显示英语简介。

• Prepare CAD assets for simulation
• Convert source assets to USD and check basic validity
• Assign materials and physics properties

Find the right workflow for next-generation sequencing data by inspecting files and clarifying only missing setup details. Receive a recommended analysis skill, checks of available tools, and plans for software installation and reference data.

中文简介正在准备中。当前显示英语简介。

• Choosing an analysis workflow from sequencing files
• Clarifying requirements for RNA-seq or variant analysis
• Checking tools and preparing an installation plan

Build or review quality control workflows for single-cell and single-nucleus RNA sequencing data. Select filters from observed data, detect doublets, assign cell types, and create UMAP plots that summarize cell relationships, while recording analysis decisions and outputs.

中文简介正在准备中。当前显示英语简介。

• Choosing quality filters from dataset distributions
• Detecting doublets and checking ambient RNA
• Assigning cell types using references or marker genes

Analyze Mixpanel data with Python to explore usage trends, conversion steps, returning users, and differences between user groups. Also manage business context documents that explain an organization's structure and goals to AI assistants.

中文简介正在准备中。当前显示英语简介。

• Find where users drop out before purchasing
• Compare retention across user groups
• Explore feature usage and revenue trends