MandateClaw is an open-core Ruby library and registry for defining, signing, and auditing programmatic contracts that govern AI agent behaviour in regulated workflows. It is aimed at engineers deploying agents in fintech, legal, healthcare, and government, where informal prompts and opaque controls are described as inadequate for compliance.
The system centers on signed contracts that are meant to be deterministic, human-readable, and auditable. Example contract syntax shows parties, obligations, permissions, prohibitions, and agent bounds, along with requirements such as signatures from multiple parties and ed25519 signing. The same definition that is enforced at runtime also renders as a human-readable contract document, and the site describes the result as readable by lawyers and auditable by regulators. The components table identifies a DSL and Markdown renderer under mandateclaw-dsl, and signing, storage, breach log, and an audit API under mandateclaw-registry.
MandateClaw is delivered as Ruby software. The component list describes mandateclaw-dsl as MIT licensed, while mandateclaw-registry is listed under FSL-1.1-Apache-2.0. The documentation also labels the overall project as open-core, and presents it as a framework for agent contracts rather than a general AI application.
MandateClaw sits in PulseGate's Frameworks & SDKs category. It focuses on enabling engineers to define, sign, and audit enforceable contracts for AI agent behavior in regulated workflows. MandateClaw is an open-source project aimed at engineers deploying AI agents. MandateClaw is open source under the MIT license. It runs on the web, and it can be self-hosted.
MandateClaw first shipped in 2026. The project is developed in the open on GitHub with 3 commits in the last 90 days. Key capabilities include contract definition, audit logging, and Ruby DSL.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do