conformance-testing

⚡️ Apache-2.0 formats for organisational accountability on the agentic web: who is accountable, which lanes an organization opens, what shipped, and what a stranger can verify offline. Every format ships a conformance corpus – most of it refusals.

TMPA — Textual Multi-Agent Process Architecture for durable AI-agent governance, responsibility, conformance, and auditable digital work. Research, specification, and evidence.

⚡️ Run a conformance corpus against any executable, in any language. One JSON object per line in, one per line out – no bindings, no SDK. Four outcomes, because a program that crashed has not answered wrongly.

The kosync protocol specification, and a verifier that proves an implementation against it.

Policy compiler and cross-client conformance lab for EIP-7906 transaction assertions

A research reference for equal-stake challenge escrow with committed terms, evidence-linked resolution, deterministic timeout exits, and reproducible assurance artifacts.

AgenticSystemCore: an open specification and reference engine that publishes knowledge Bundles for people and agents — RFC 9264 discovery, a JSON-LD graph, llms.txt and MCP tools

A reproducible conformance suite for Confluence to Markdown exporters — 56 rules, a seed corpus, and a scorer.

CAPProbe - a paid CROO agent that conformance-tests other CAP agents. CROO Agent Hackathon.

Behavioral conformance evidence for Jev integrations: native host tests, finite outcomes and provenance

Independent conformance kit for x402-foundation/x402#3220 (authority mandate) and #3376 (authorization-evidence EVC boundary). Zero dependencies, 39/39 vectors.

A dependency-free conformance harness for the x402 v2 payment protocol — with the exactly-once tests the reference SDKs were missing.

Evidence-first GitHub Action and CLI for verifying released Codex plugin discovery.

Target-agnostic Artillery load & conformance tests for S3 static-website semantics (native S3 or s3proxy)

Verify a sealed session record: hash chain, Ed25519 signatures, time receipts, and audio anchors. Library, CLI, and browser verifier in one package.

Implementation-independent conformance evidence for agent-run integrity.

Wornpage validation toolkit: check page-owned WebMCP tools, authority boundaries, lifecycle contracts, and receipts.

LiveView interaction contracts: a conformance harness + delegation ledger for browser-native interaction state under Phoenix LiveView patching (not a component library).

Adversarial conformance testing for agent runtimes, action boundaries, transports, and committed effects.

Cut a stream at every point and check what the consumer is left with