Agent harness for long-running AI coding tasks — orchestrates Claude Code & GitHub Copilot across repositories 🍩

A framework for executing long-running AI agent tasks with structured feature lists and progress tracking.

Java-native runtime for long-running AI agents with live execution control, human approvals, and audit trails.

Long-running AI agents that adapt until outcomes converge.

Production 3-agent harness for autonomous multi-hour software builds. Planner + Generator + Evaluator with parallel fleet execution, circuit-breaker stagnation detection, and graceful resume.

watches the watchers: a monitoring system for long-running Claude Code agents

An autonomous AI coding agent for long-term task execution.

Codex-derived source release for an opt-in session compact route that preserves durable handoff state across repeated compactions.

Architecture notes on CLI-native autonomous coding agents, tool runtimes, orchestration, UI surfaces, and protocols.

Codex skill for designing and scaffolding durable agent harness projects, with a Ralph Loop preset and upgrade-friendly doctrine.

Production-ready workspace setup for long-running OpenClaw AI agents. Artifact workflow, secrets management, memory optimization, and battle-tested patterns.

Claude Code plugin for long-running agent workflows with progress tracking, feature checklists, and git checkpoints

Evidence-bound workflow diagnostics and certified lower-bound reporting for long-running AI agent pipelines. Improve agent workflow throughput without changing the model.

Continuity for AI agent work that outgrows one chat.

Personal website of Iris Shen, focused on memory, evaluation, orchestration, and runtime systems for long-running AI agents.

Observable-only workflow memory for long-running agents: promotes raw short-term traces into verified, receipt-bound workflow memory without relying on hidden meta-evaluators.

Theory-to-experiment lab for search stability in long-running agents under finite context, with exact simulator tests and lightweight mechanistic probe tasks.

Certified Memory Governance Layer for long-running AI agents: strict receipts, append-only ledgers, authority gates, retrieval filtering, telemetry replay, and safe adapters for Mem0, Graphiti, LangMem, and LangGraph.

Local-first, model-agnostic workflow optimizer for long-running AI agents: observable JSONL ledgers, deterministic reducers, no-meta gates, and receipt-backed self-improvement without LLM judges or model-weight updates.

🤖 Agent Skills compliant skill for autonomous AI agents that parse PRDs into tasks and execute them. Works with Cursor, OpenCode, Claude & any AI framework. Features state persistence, dependency management, error recovery.