Analyze datasets by running clustering algorithms (K-means, DBSCAN, hierarchical) to identify data groups. Use when requesting "run clustering", "cluster analysis", or "group data points". Trigger with relevant phrases based on skill purpose.

Build automated machine learning pipelines with feature engineering, model selection, and hyperparameter tuning. Use when automating ML workflows from data preparation through model deployment. Trigger with phrases like "build automl pipeline", "automate ml workflow", or "create automated training pipeline".

Process identify anomalies and outliers in datasets using machine learning algorithms. Use when analyzing data for unusual patterns, outliers, or unexpected deviations from normal behavior. Trigger with phrases like "detect anomalies", "find outliers", or "identify unusual patterns".

Execute orchestrate multi-agent systems with handoffs, routing, and workflows across AI providers. Use when building complex AI systems requiring agent collaboration, task delegation, or workflow coordination. Trigger with phrases like "create multi-agent system", "orchestrate agents", or "coordinate agent workflows".

Validate AI/ML models and datasets for bias, fairness, and ethical concerns. Use when auditing AI systems for ethical compliance, fairness assessment, or bias detection. Trigger with phrases like "evaluate model fairness", "check for bias", or "validate AI ethics".

Build and evaluate classification models for supervised learning tasks with labeled data. Use when requesting "build a classifier", "create classification model", or "train classifier". Trigger with relevant phrases based on skill purpose.

Process images using object detection, classification, and segmentation. Use when requesting "analyze image", "object detection", "image classification", or "computer vision". Trigger with relevant phrases based on skill purpose.

Re-derive something written into the shortest version that still covers everything. Use when something already written should get shorter and nothing it covers can be lost.

One committed verdict on something that exists, cited. Use when something exists and a claim about it needs a verdict.

Find the route to guaranteed failure, negate it. Use when a goal keeps not being reached head-on.

Put one settled thing in the user's head, plainly. Use when the user does not hold something already settled.

Draw out what the user holds but can't say. Use when the user knows something they cannot yet express.

Build the thing the user can't specify, choices surfaced. Use when the user needs something to react to before they can choose.

Plan the skills an ask needs, invoke each as its condition arrives. Use when the ask has multiple aspects that different skills carry, or the user named multiple skills.

Turn words with two readings into one confirmed ask. Use when the user's words admit two readings that would build different things.

Rewrite a rule so anyone applies it the same way. Use when a stated rule or claim takes the reader's judgment to apply.

Use when adding or changing a skill, hook, or runtime file in this repo — including deciding which unit a new capability should be.

Run the upstream Flux controller patch release procedure for helm-controller, image-automation-controller, image-reflector-controller, kustomize-controller, notification-controller, source-controller, and source-watcher. Use when preparing a new controller patch release from a release series branch, drafting changelog entries, tagging releases, and opening the follow-up changelog PRs back to main.

What it does: Chainlink Confidential AI Attester (alpha) runs an LLM inside AWS Nitro Enclave/TEE over private documents and returns the model result; raw documents do not leave TEE. When to use: trigger on private inference, attested AI, TEE inference, confidential AI, or sensitive-document analysis, including lending, accredited-investor, KYC/AML, and proof-of-reserves. Key capabilities: submit document resources, choose model, request structured JSON, poll completion, and use the result without exposing source documents onchain. Do not use for generic TEE/enclave work without document inference; use CRE Confidential Workflows for confidential Chainlink workflow logic. Do not trigger on generic confidentiality/TEE/enclave requests that involve no document upload or private inference — a user who wants a Chainlink workflow's own logic or data kept confidential from node operators, with no document analysis involved, wants CRE Confidential Workflows, which chainlink-cre-skill covers.

Creates the plugin's own Python environment and installs or upgrades mixpanel_headless (0.3.0 or newer) with pandas, numpy, matplotlib, seaborn, networkx, anytree, scipy, and pyarrow (Python 3.11+ only) on Python 3.10+, then checks the imports, mp help, and the credentials. Use when setting up Mixpanel analysis, or when the plugin environment is missing or older than 0.3.0. Do not use for login or account changes (use auth).