Signatrust is a platform designed to provide cryptographically signed, independently verifiable receipts for business-critical decisions made by AI agents. Addressing the challenge of proving and auditing AI-driven actions, it enables organizations to create tamper-evident records—called Decision Receipts—each sealed with Ed25519 signatures and linked in an append-only ledger. This approach allows auditors, insurers, and regulators to verify what an AI agent did, who authorized the action, and which policies governed it, all without accessing the underlying raw data.
The platform supports a full trust chain, encompassing the signing of AI decisions, independent verification of receipts, generation of per-agent trust scores, and aggregation of risk profiles for insurance purposes. Features include the ability to register AI agents, declare their models and permissions, manage API keys and policies, and hash documents in-browser for zero-custody signing. The open specification for Decision Receipts (ADR Spec) and the canonical receipt format enable third-party vendors to implement compatible solutions. Trust Passports provide a reputation profile for each agent, and trust scores are computed from verifiable histories rather than self-attested claims. Risk scores are aggregated across an organization's agents, supporting insurability and underwriting based on real evidence.
Signatrust integrates with workflow automation via a verified n8n Node, allowing users to sign every workflow decision within n8n. The platform offers a REST API, SDKs, and open-source samples, and its products are built on a unified, append-only ledger. Security is emphasized by ensuring that decision data never leaves the user's perimeter, with all input and output fingerprinted using SHA-256 hashes on the client side.
The platform is positioned for use cases in compliance, audit, risk management, and regulatory readiness, including preparation for upcoming regulations such as the EU AI Act. It is relevant for enterprises, auditors, insurers, and regulators seeking audit-grade, independently verifiable trust layers for AI-driven processes.
In the LLM eval & observability space, Signatrust takes a focused approach. It focuses on ensuring AI agent decisions are provable, auditable, and insurable through cryptographically signed receipts and trust scoring. It is built as a B2B product for AI developers and organizations deploying autonomous agents. A free plan is available. It runs on the web and API.
Signatrust builds and maintains Signatrust, and the product first shipped in 2026. Development happens publicly on GitHub with 21 commits in the last 90 days. PulseGate's similarity index finds few close equivalents — Signatrust occupies a relatively distinct niche. Key capabilities include decision receipts, cryptographic signing, and receipt verification. It exposes integrations via a public API.
Latest indexed changes and source events
Do AI agents need cryptographically verifiable decision receipts? discovered by the PulseGate indexer
Other apps tracked under the same category.
