indirect-injection-detection
Detects indirect prompt injection and goal hijacking in AI agents by evaluating how they process external content like RAG, documents, and web data.
Discover reusable agent skills, browse implementation details, and find the right skill for your workflow.
486 skills found
Detects indirect prompt injection and goal hijacking in AI agents by evaluating how they process external content like RAG, documents, and web data.
Convert PRDs, API docs, and requirements into structured acceptance, testing, integration, and launch checklists.
Design and document REST or GraphQL APIs, including endpoint definitions, pagination, filtering, versioning, and OpenAPI/Swagger specifications.
Automate WordPress content publishing with draft workflows, media library integration, and native Hebrew/RTL support.
Automates the end-to-end release process for R3BL workspace crates, including version bumping, changelog updates, testing, and publishing to crates.io.
Manage isolated LlamaFarm development environments using git worktrees for parallel agent sessions and service testing.
High-performance build system for JavaScript/TypeScript monorepos. Optimized for parallel task execution, remote caching, and dependency graph analysis.
Generates a random lucky number between 0 and 9999 for games, decision-making, or entertainment.
Senior backend architecture expert specializing in Hexagonal Architecture, DDD, SOLID principles, clean code, and refactoring to guide development, reviews, and architectural problem-solving.
Keep your technical specifications, test suites, and source code perfectly synchronized during AI-assisted development.
Implement a full Model Context Protocol (MCP) stack in Rails. Connect to external servers, expose your Rails app as an MCP server, or manage subprocess MCP containers via Docker with OAuth 2.1 PKCE support.
Generates standardized metadata, including git/jj version info and timestamps, for research docs, handoffs, and implementation plans.