🎩 You're Invited:Meet the Socket team at Black Hat in Las Vegas, August 3-6.RSVP
Sign In
e

ecolibria

npm

6.18K

total weekly downloads

Packages

Current Co-maintainers

Former Co-maintainers

43 packages

ai-trust

Trust verification CLI for AI packages — check MCP servers, A2A agents, AI tools, and LLMs before you install

arp-guard

Agent Runtime Protection — LLM-assisted behavioral monitoring, anomaly detection, and enforcement for AI agents.

cryptoserve

CryptoServe CLI - Cryptographic scanning, PQC analysis, encryption, and local key management

damn-vulnerable-ai-agent

The AI agent you're supposed to break. 17 agents, 12 vulnerability categories, zero consequences.

hackmyagent

Find it. Break it. Fix it. The hacker's toolkit for AI agents.

hackmyagent-core

Core security scanning library for AI agents

@nanomind/atc

ATC intent handler — trust queries against the OpenA2A Registry

@nanomind/cli

NanoMind unified CLI entry point — interactive AI-powered security assistant

@nanomind/daemon

Persistent NanoMind inference daemon with HTTP and IPC interfaces

@nanomind/engine

Core inference backend — llamafile wrapper for local LLM inference

@nanomind/guard

Prompt injection detection for CLI input

@nanomind/hma-adapter

NanoMind adapter for HackMyAgent — interactive mode when hma runs with no args

@nanomind/opena2a-adapter

NanoMind adapter for OpenA2A CLI — cross-product interactive router

@nanomind/router

Intent classification and command mapping for CLI tools

@nanomind/runtime

NanoMind Runtime — behavioral twin for ARP anomaly detection

@nanomind/runtime-core

NanoMind RuntimeTwin: behavioral twin (LSTM) for ARP anomaly detection. Canonical home of the LSTM scorer referenced in NANOMIND_V3_AUDIT.md Section 9 item 1.

@nanomind/secretless-adapter

NanoMind adapter for secretless-ai CLI

@opena2a/a2a-idf

Reference TypeScript SDK for A2A-IDF (Agent-to-Agent Identity Framework) — sign/verify RFC 9421, resolve keyids, validate verification levels 0/1/2, attestations, and delegation chains.

@opena2a/ai-classifier

Tier-based AI package classification (native / adjacent / unrelated) for OpenA2A trust tooling

@opena2a/aicomply

Inline PII, credential, and regulated-data classifier for AI agent I/O. Dual-layer: deterministic regex + optional semantic Guard. Library + CLI.

@opena2a/aim-core

Lightweight agent identity library — Ed25519 identity, local audit log, capability policy, trust scoring. No server required.

@opena2a/aim-sdk

Agent Identity Management SDK for TypeScript/Node.js

@opena2a/arp

Agent Runtime Protection — LLM-assisted behavioral monitoring, anomaly detection, and enforcement for AI agents.

@opena2a/atx-verify

Spec-compliant offline verifier for ATX (Agent Trust eXtension) credentials — Ed25519 signature verification over the canonical payload (v1.0 pipe / v1.1 JCS RFC 8785), expiry, revocation, issuer-trust, and issuer-chain checks. Byte-for-byte interoperable

@opena2a/check-core

Data-shape and orchestration primitives for the `check` command across OpenA2A CLIs. One implementation of input classification, registry → scan-on-miss flow, download-error translation, and canonical CheckOutput schema — consumed by hackmyagent, opena2a-