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.
123 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.
Extract plain text from EPUB, MOBI, and PDF files for analysis or processing. Includes local support for all common ebook formats.
Retrieve current, source-backed technical information using MCP tools to resolve queries about libraries, APIs, SDKs, and evolving tech ecosystems.
Audit, prune, and maintain vector memory for Clawdbot. Prevents token waste, clears junk data, and automates memory hygiene via LanceDB maintenance.
Autonomous research specialist for verified information gathering, source evaluation, and structured synthesis.
Handles large-scale tasks by automatically breaking them down into manageable, recursive sub-tasks to overcome context window limits and improve reasoning accuracy on large codebases and document sets.
Official Mastra framework guide. Master AI agent and workflow development with local documentation lookup, API verification, and TypeScript-based project management.
High-performance document intelligence library for extracting text, tables, code, and metadata from 91+ file formats, with OCR and LLM-ready output.
Advanced Gemini-powered web search plugin with smart caching, subagent context isolation, and automated query optimization.
Execute z.AI CLI for multimodal analysis, web search, reader, and GitHub repo exploration via CLI and MCP.
Research agent for Nia: index/search remote codebases, docs, and packages. Optimizes AI context by prioritizing full source indexing over web fetches to reduce hallucinations.
Perform internet searches using the Zhipu AI web search API to retrieve real-time information, news, and current data.