An AI-powered brand governance platform that reviews content against organizational guidelines using hybrid RAG, multi-agent workflows, and evidence-based validation.
The native UI/UX protocol for AI agents. SKILLS
作物用药场景的规则门禁与离线评测工程原型；安全结论仅限给定测试范围。
Offline-first security alert triage CLI with deterministic rules (authentication, process, network), an escalation-only scikit-learn advisory classifier, and a prompt-injection-safe local Ollama LLM summarizer.
Vision-language agent that verifies damage claims from customer photos across cars, laptops, and packages. Deterministic rule layer ensures safe decisions and escalates weak evidence.
Works out which EU AI Act risk tier an AI system falls into and lists the obligations that follow, each with its article. A lawyer confirms the result.
Auditable AI-assisted credit monitoring: deterministic rules decide, LLMs read and quote, every claim sourced
A demo for small accounting firms that reads each invoice twice (plain code and AI), checks the result with plain rules, and routes anything unclear to a person.
Turns a public-domain or openly licensed ruleset into a deterministic rules engine, with provenance tying every line back to its source.
Ruleset-agnostic determinism primitives for rules engines: replay identity, provenance, and the unresolved-result contract.
Checks a draft crypto-asset white paper against the Parts of MiCAR Annexes I to III and reports each candidate gap with its citation. Browser playground, CLI and GitHub Action.
Billing dispute demonstrator on synthetic data: deterministic checks first, an LLM interprets complaints, code calculates correction amounts, and humans approve financial actions. Auditable, provider- and model-agnostic. Built with spec-driven development and cross-vendor code review.
Build design structures for AI coding agents with a structured framework for complex project architectures.
To associate your repository with the deterministic-rules topic, visit your repo's landing page and select "manage topics."