Skill UI

Checks website, GitHub, and local docs to identify messaging inconsistencies and captures version, feature, terminology, and contact differences in a read-only discrepancy report whenever a consistency check is requested.

Analyzes smart contract codebases to enumerate externally callable state-changing functions, classify their access (public/role/contract-only), and produce structured audit-ready reports for Solidity, Vyper, Solana/Rust, Move, TON, and CosmWasm projects.

This skill provides comprehensive enterprise-grade access control (RBAC) for Instantly.ai workspaces. It enables administrators to manage workspace members, assign roles, govern API keys with granular scopes (e.g., read-only, specific campaign access), and set up workspace groups for agency operations. Essential for implementing secure, multi-user environments and maintaining strict data isolation.

A comprehensive guide detailing security best practices for integrating with Lokalise. Covers strategies for managing API token scopes (read-only vs. read-write), enforcing the principle of least privilege, validating translated content for XSS or credential leakage, verifying webhook secrets, and securing CI/CD pipelines against hardcoded secrets. Essential for developers and DevOps engineers.

This skill provides comprehensive, structured audit logging for OpenRouter API interactions. It automatically captures essential metadata like total cost, token usage, model details, latency, and status, while also performing PII redaction. The resulting audit trail is stored in an append-only SQLite database, making it ideal for building compliance-ready systems, detailed cost monitoring, and deep debugging of production environments.

Read-only assistant that inspects Sentry issues, summarizes recent production errors, and pulls health metrics via the Sentry API when users ask for error reports or recent crashes.

Analyzes a code repository to generate a recommended, minimal set of read-only shell command permissions for Claude Code's settings.json. It automatically detects the project's tech stack (e.g., Python, Node.js, Go), build tools, and monorepo structure, ensuring a safe and evidence-based security baseline for new projects or auditing existing permissions.

Use this skill to perform read-only API calls for exploring, previewing, searching, and filtering structured datasets hosted on Hugging Face. It allows users to validate dataset availability, list splits, paginate large datasets, and retrieve comprehensive metadata and statistics, making it ideal for data validation and content extraction workflows.

This comprehensive index provides access to 32 production-ready, specialized engineering skills. It covers the entire development lifecycle, including system architecture, frontend, backend, DevOps, security operations, AI/ML, and data engineering. It is designed to guide users in selecting and loading only the specific module needed, ensuring focused expertise for complex tasks.

Evaluate page content for usefulness, E-E-A-T signals, readability, thinness, and AI citation readiness. Includes last-mile draft cleanup such as AI-typical phrasing and invisible Unicode watermarks. Content-only analysis, not full-page technical checks.

This skill enforces a mandatory, structured gate for starting new implementation tasks. It mandates that all work must be tracked via a GitHub issue, requiring explicit, testable acceptance criteria before any execution can begin. It defines a strict workflow (Problem, Goal, Scope, Acceptance Criteria, Status) to manage task readiness, ensuring that development only proceeds when criteria are fully defined and verifiable.

Validates SKILL.md files against the agentskills specification and Anthropic best practices, catching structure, semantic, naming, and quality issues in a single read-only pass before publishing or debugging skills.