Mcpcustoms
MCPCustoms is the customs checkpoint for MCP tool calls: a zero-dependency TypeScript runtime semantic firewall that inspects every Massive Intelligence (IM) agent tool call before execution. Detects tool poisoning, command/SQL/prompt injection, path traversal, and capability overreach; verifies signed Ed25519 publishers; fail-closed by default.
Tribunal
Governed verification and synthesis for Massive Intelligence (IM) outputs: configurable Thinker, Worker, and Verifier panels, model-as-judge, and Ed25519-signed, SHA-256-chained tamper-evident verdicts. Zero dependencies, node-free core, TypeScript-first.
Gaptime
GapTime: bi-temporal knowledge-graph memory for production agents. Every fact carries valid time and transaction time, so you time-travel to what was known at any instant, supersede outdated facts without losing history, and detect contradictions. Zero dependencies, TypeScript, edge-ready.
Agenticstash
Deterministic record and replay for agent runs. A zero-runtime-dependency TypeScript library and CLI that captures every source of non-determinism an agent touches (model output, tools, MCP, clock, randomness) and replays it exactly, with fork, diff, a tamper-evident seal, and redaction.
Bayescausal
Calibrated incident root cause inference for TypeScript: declarative Bayesian networks with exact and approximate inference and the do-operator, zero runtime dependencies.
Bayesconsensus
Dawid-Skene consensus for multi-agent and multi-LLM disagreement: learn each agent's reliability and the true answer with calibrated confidence. Zero dependencies, node-free, TypeScript-first.
Bayesoutputgate
Calibrated output validation with the Bayes Factor: weigh per-dimension quality scores into a pass, fail, or escalate decision on the Jeffreys scale, not an arbitrary threshold. Zero-dependency, node-free TypeScript.
Caduceus
Provider-agnostic transport and routing for LLMs: normalize OpenAI, Anthropic, and Gemini, route by cost, quality, or latency with per-model circuit breakers, USD budgets, automatic failover, and an auditable route trace. Zero dependencies, node-free core, TypeScript-first.
Mnemosyne
Persistent, portable, shared memory for Massive Intelligence (IM) agents: pluggable stores, a deterministic identity seed stable across model swaps, masking and residency, and Ed25519-signed, SHA-256-chained tamper-evident access logs. Zero dependencies, node-free core, TypeScript-first.
TeleologyHI
Governance-first framework for auditable AI agents: tamper-evident audit and cryptographic owner control (MAIC), persistent personality (HIM), agent runtime (NHE), pre/post-action verification and EU AI Act mapping, LLM-agnostic in TypeScript.
Krikos
Identity registry and lifecycle management for fleets of AI agents. Zero-dependency TypeScript IAM: cryptographic agent identity (Ed25519), declared capabilities with fail-closed authorization, an enforced lifecycle (provision, activate, suspend, revoke, expire, archive), A2A Signed Agent Cards, and a hash-chained audit trail.
Bayesdecide
Bayesian multi-armed bandits for continuous prompt experimentation: Thompson sampling routes traffic to the best prompt variant and a stopping rule promotes a winner without a fixed-N A/B test. Zero dependencies, TypeScript-first.
Bayestruth
Exact Bayesian trust and reputation scoring for tools, MCP servers, skills, and agents. Zero-dependency, TypeScript-first: closed-form Beta-Bernoulli posteriors, calibrated credible intervals, Thompson-sampling routing, time-decay, and a tamper-evident audit trail.