Self-Improving Agents -- A Progression Four levels of self-improving code agents, from the simplest loop to a full adversarial arena with self-modifying agents. Each level adds one key idea.

Continual agent skill evolution through persistent decision history. Whole-skill optimisation (SKILL.md + scripts + references) with every decision landing as a local Git issue / PR / wiki. Runs on any agentskills.io runtime — Claude Code, Codex, OpenClaw, Hermes.

Agent skill for running Codex or Claude Code as an orchestrator over Symphony workers and Linear issues. Plans waves, dispatches workers, reviews and merges, and optionally pursues a goal across many waves under hard budget caps.

A lightweight, declarative agent harness — define multi-agent workflows as YAML, run them from Python or the CLI, and they get measurably better every run.

Standalone self-improvement harness for agent traces, memory, skills, evaluation, and fine-tuning exports

Memory that learns and keeps itself current. A six-layer memory stack for Claude Code plus a nightly learning loop (capture, consolidation, scouts, conductor) that promotes your lessons into rules and surfaces new tools that fit your stack. Free, MIT.

Shogun AFM is Agent Fleet Management for self-improving AI agents — combining agent orchestration, persistent memory, fleet monitoring, governance, security posture, and Gensui command control.

AI agent workspace architecture, demonstrated end-to-end in Claude Code: roles library, persistent memory, hooks, scheduled agents, self-audits, loop selection, and measurement-gated self-improvement. Interactive tour, fork-ready samples.

repo for reusable plugins and skills

Beastmode: MofA (Mixture of Agents) orchestration framework for Hermes/OpenClaw/Codex with MemroOS-style context continuity.

Self-improving repo health remediation skill with audit/fix/diff modes, evidence coverage, counter-review, and stable HP-* findings.

Local-first MemoryOps skill for AI agents: diagnose, patch, eval, and evolve memory safely.

ACE — Agentic Context Engineering: evolving, self-improving context playbooks for LLM agents. Faithful ICLR 2026 implementation with OpenAI Agents SDK support.

Loop engineering plugin for Claude Code — persistent memory vault, self-correction hooks, and a 5-stage failure-to-knowledge distillation protocol. Self-improvement as a system, not a model.

Give your AI agent memory. Convenience wrapper for agent-episodic-memory.

Skill Forge — turn an LLM agent's repeated failures into reviewed, signed-installation skills. Telegram-approved HMAC tokens, replay regression gate, and a full Forge Console covering doctor → demo → forge → install → evolve.

Tirami — distributed LLM inference where compute is currency. 1 TRM = 10^9 FLOPs. 21B supply cap, yield halving, staking, collusion resistance. 100% Rust, OpenAI-compatible, no token, no ICO. "tira mi su" = pull me up.

Five runnable demos for building resilient AI agents with Strands Agents: chaos-test the agent's tools, gate memory writes to stop hallucinations, defend against memory poisoning and prompt injection, verify multi-step tasks against the backend, and let an agent write its own tools. Runs on OpenAI or Amazon Bedrock.

Full self-improving long-horizon LLM agent with strategy memory, failure analysis, Grok teacher labels, and QLoRA student distillation.

Methodology skill + deterministic harness for agents that self-iterate any skill/repo/project. Anti-self-deception by design — an un-gameable acceptance gate so 'accepted = real improvement', not a fake score-up curve. Built for the open-ended, no-ground-truth domains the self-improving-agent literature skips. 521 tests.