jovd83/lifelike-synthetic-data-generator
Generates realistic synthetic datasets for tests and demos with seeded, locale-aware, and distribution-shaped data.
Generates realistic synthetic datasets for tests and demos with seeded, locale-aware, and distribution-shaped data.
npx add-skill jovd83/lifelike-synthetic-data-generatorGenerates realistic synthetic datasets for tests and demos with seeded, locale-aware, and distribution-shaped data.
Research pipelines as semantic execution units: each skill declares inputs/outputs, acceptance criteria, and guardrails. Evidence-first methodology prevents hollow writing through structured intermediate artifacts.
Trigger-Driven Dynamic Context Loading for Code-Aware LLM Agents
Reduce cloud LLM costs by 40% with local context indexing. Use precision retrieval tools for MCP-compatible agents to minimize token usage and process data locally.
“VCR for agent tools” / Agent Replay
Optimize LLM agents with train-free Kode forward runs, skill evolution, and beam search prompt pruning for better performance
Security audit skill for Claude Code community skills. Scans for prompt injection, data exfiltration, permission bypass, dangerous commands, and supply chain risks.