A

ai-agents

An autonomous agent library for C++23. It ships as libdaedalus.a plus four worked example hosts (demos/). Embed it in any C++23 host (CLI, web, IDE plugin, CI bot) and drive one or more Agent instances via the public API. See docs/embedding.md to get started.

Takes a private repo public without a public mistake: audit, secret scan, governance files, README polish, and release prep, with every step stopping at a reviewable result before anything goes outward.

Agent UI design with hard gates: generate, render in a real browser, judge against accessibility and an honesty floor (unknown data shows as unknown), repair, and repeat until it passes.

Authors and tests the small skill files agents run on: one measurable outcome per skill, its own evals, and packaging generated for both Claude Code and Codex. Nothing installs or promotes itself.

Research that argues with itself: an adversarial refuter attacks every draft, typed claims are verified against their cited sources, and the engine abstains honestly when the evidence is thin.

One local agent hands a bounded task to another: Claude Code and Codex work through their own logged-in clients, no shared keys. Fail-closed launchers, a receipt for every run, results come back reviewable.

Hand an agent work too big for one sitting and walk away: queues with spending caps, checkpoints, resume after limit resets, and a reviewable morning report. Bundles Claude Carry, the overnight runner.

Picks the single best capability for a task (skill, tool, MCP server, integration), installed or not: better options surface as owner-approved choices. Deterministic receipts, refusals with named reasons, never runs anything itself.

Cognitive exoskeleton for AI agents and human researchers — by Cognitive Functors. Turn questions into verified constructive & destructive insights: breakthrough scientific-engineering hypotheses checked across 47 literature sources, 38 simulation bridges, and 9 formal verifiers — then gated proposals, dissertations, BibTeX, and proofs you can publish. Not a chatbot — a research pipeline. pip install c4reqber · blast solve · Z₃³ · 21 MCP · AGPL-3.0.

A shared, git-backed knowledge base and message bus for AI coding agents — built on nothing but a GitLab wiki. Templates, conventions, a Claude Code skill, and a maintainer/lint kit. Born at gethal.ai.

One dashboard to manage every Claude Code agent on your machine: spawn in git worktrees, live kanban + orchestrator/subagent trees, permission bridge, steer/kill/merge, and background-bot heartbeats.

A Coinbase x402 micropayment-gated API proxy for The Undesirables TCG Oracle. Allows AI agents to semantically discover and autonomously pay for physical trading card grading and Monte Carlo market intelligence using USDC on the Base network. Fully integrated with the CDP Bazaar Discovery layer.

Solble, by Droppy. Claude directs, GPT builds, one chat. Free to use.

icarus is a CLI AI coding agent built on the daedalus framework. It is an autonomous coding agent that can explore repositories, read and write files, run shell commands, and call tools — either interactively from the terminal or as a subprocess driven by an IDE.

Before/after screenshots for GitLab UI merge requests: captured locally in a GDK checkout, verified, attached to the MR. Agent-agnostic (AGENTS.md); installable as a Claude Code plugin; no dependencies beyond the gitlab repo itself.

LLM-powered Minecraft AI agents (Mineflayer), modernized for Node.js 24 LTS — prebuilt native deps, npm 12 ready, CI, zero lint errors. Fork of mindcraft-bots/mindcraft.

Production-grade memory operating system for agentic AI. Interoperates with every major framework via MCP, OpenAI-compatible gateway, and native /v1/* REST — Claude Code, OpenClaw, ZeroClaw, Hermes, MemPalace, LangChain, LlamaIndex, CrewAI, AutoGen, Mem0, Letta, Zep. In production use since December 2025. Apache 2.0.

Encrypted reminders, scheduling, tags, SSE, and natural language agent API.

https://demsking.gitlab.io/oremi-isinmi-agent

Thin TypeScript + zero-dep Python client and recipes to gate high-risk actions behind a payload-bound passkey approval.

ComparEdge MCP server: verified pricing, plans, ratings and alternatives for 494 software and AI tools