Search

Deterministic Agentic Governance Framework for Autonomous Software Engineering

Open-source in-process governance for AI agent frameworks: Cedar-governed model/tool calls via build_middleware()/GovernedAgent, plus the Cedar engine, request/decision shapes, PEP<->control-plane protocol client, and Ed25519 PEP identity they run on.

HeartFlow (心虫) discriminator wrapper for LangChain — guard LLM input/output before it reaches users.

Coordinator that runs a small graph of AI-agent nodes with bounded spend, mechanical gates and journaled state

Deterministic policy engine for LLM-generated SQL: SELECT-only, PII column denylist, table allowlist, cost cap. Multi-dialect via sqlglot (BigQuery, Snowflake, Postgres, Trino, …).

A strictly-typed, provider-agnostic AI Agent SDK

The canonical read-only local command-line composition for moldea repositories.

Vendors shared reference documents into agent skill directories and locks what each skill depends on.

Official TypeScript/JavaScript SDK for the CANDY AI Marketplace — deploy an AI business in 3 lines.

Give it a store URL — it detects the platform and, for custom sites, synthesizes a reusable scraper once via an LLM, then replays it for free.

Local declarative page builder for agent-authored interactive HTML artifacts.

Cost-aware dynamic-topology multi-agent LLM orchestration (DyTopo-based)

LangChain middleware for the EISFI Weaver integrity chain: a tool call whose integrity precondition fails is never executed.

Deploy a team of chat-resident AI agents from declarative manifests: Hermes + litellm + acp2api + subscription coding CLIs, one container per agent.

AI-powered Jira CLI assistant

Tool registry and execution pipeline for the DeepSeek Harness

LangChain and LangGraph deterministic invariant tool guard

System prompt assembly registry for the xneog

The default executor-less/UI-less agent spine with fallback session titles, provider-routed retry, and optional persisted goals

ANTRI Code - The Autonomous Agentic CLI & Desktop Assistant