3K skills · sorted by relevance
Organize PyTorch training code with LightningModule and Trainer. This skill guides training, validation, logging, model checkpoint saving, early stopping when progress stalls, and configuration for running the same model across multiple GPUs.
Build NestJS APIs with TypeScript code for modules, request handlers, services, authentication, input validation, and database access. The skill also prepares Swagger API documentation and unit and end-to-end tests for the implemented features.
Build machine learning workflows that connect data checks, training, evaluation, and deployment. The skill implements experiment tracking, model version management, and automated retraining, with reproducible environments and explicit validation thresholds before models enter production.
Inspect repository code or code changes for plausible security vulnerabilities. The skill records candidate issues with affected locations, attacker-controlled inputs, failing safeguards, and possible impact, preserving the evidence needed for later validation.
Assess candidate security findings using focused reproduction, tests, or code tracing. Record the evidence, confidence, and remaining proof gaps for each candidate, helping determine which findings warrant reporting and which remain uncertain.
Prepare a structured JSON handoff for feature work, bringing together requirements, design decisions, key user journeys, tests, demo recordings, and pull request status. Clearly distinguish completed work from missing validation, publication steps, and other remaining tasks.
Plan, run, and merge several small development tasks as one batch in Pi. A coordinator manages shared files and worktrees, while options control concurrency, add research and validation, or resume eligible tasks from an existing batch.
Run several small development tasks as one batch, from planning through execution and merging. Control concurrency, add research or validation, and resume eligible unfinished items from a saved batch.
Review code in a convex/ directory for security, access control, input and output validation, performance, and schema design. Receive findings grouped by severity, with explanations of each issue and suggested fixes before release.
Evaluate a feature idea before turning it into requirements using six sequential questions. Examine customer demand, current workarounds, a specific user's pain, and the smallest useful solution, then save an evidence-based assessment with a verdict and next steps.
Define reusable OpenRig agent setups, roles, and workflows in specification files. Review dependencies that prevent use on another host, distinguish specifications from bundles and extensions, and verify both file structure and actual execution.
Guide an approved defect fix from reproduction on current main through testing and independent validation. Record evidence for each operator workflow, track change size and rollback scope, and apply review records when receipt-driven development is enabled.
Build dashboard panels in plain TypeScript using a shared Panel class. Examples cover form submission, input validation, error messages, browser storage, and immediate display updates, with a finance tracker demonstrating how the pieces fit together.
Find input files in confirmed workspace folders, save outputs to explicit locations, and check the results against task requirements. Useful in restricted execution environments where familiar folders may be missing or inaccessible.
Assess a single suspected security issue using Trailmark's graph of code relationships. Gather evidence about reachable paths, input flow, and potential impact, then receive a priority verdict and a list of remaining manual checks.
Reviews Algorand smart contracts written in TEAL or PyTeal for 11 vulnerability patterns. Produces a complete coverage table and findings with code references, severity ratings, and suggested fixes, with optional automated checks using Tealer.
Review Solana and Anchor programs for six security patterns, including unsafe calls to other programs and missing account checks. Receive a complete coverage table, findings with file references and severity, and suggested fixes for identified issues.
Validate security patches by comparing vulnerable and patched code against exploit tests, related attack variants, and normal behavior. Produce reproducible logs showing remaining flaws, regressions, newly introduced security failures, and gaps that still need testing.
Assess JWT authentication in web and mobile apps using a structured security checklist. Examine token signatures, key handling, validation, and storage to investigate authentication bypass risks and identify relevant fixes.
Build an initial customer profile from existing research, market data, and team knowledge. Organize behaviors, frustrations, goals, and buying influences into a shared working hypothesis, with assumptions clearly marked for later validation.
Set up NVIDIA infrastructure for synthetic data generation on MicroK8s or Azure AKS. Coordinate Kubernetes, inference services, and OSMO workflow execution, with prerequisite checks, deployment validation, capacity planning, and failure investigation before workloads are retried.
Examines code or code changes for plausible security vulnerabilities by tracing attacker-controlled input to dangerous operations or failed safeguards. Records candidate issues with affected locations, supporting evidence, potential impact, and controls that need further validation.
Assess candidate security findings using focused reproduction, tests, or code tracing. Record the evidence, confidence, and remaining proof gaps for each candidate, helping determine which findings warrant reporting and which remain uncertain.
Add native macOS behavior when SwiftUI alone falls short. Choose a small AppKit bridge for views, windows, panels, or menus, while clarifying state ownership, data flow, and lifecycle risks that need validation.