AcontextSkill memory layer for agents: auto-captures learnings from runs into plain Markdown skill files you can read, edit, git and share across frameworks — memory without an opaque store.
Why switchBoth auto-capture learnings from agent runs into plain Markdown skill files; acontext is framework-agnostic, autoharness a Claude Code plugin that also merges and prunes its library.
Full comparison → claude-reflectClaude Code plugin that learns from your corrections — hooks capture them in-session, /reflect syncs approved learnings to CLAUDE.md/AGENTS.md, /reflect-skills mines history into reusable commands.
Why switchBoth learn from your Claude Code sessions; claude-reflect syncs approved corrections into CLAUDE.md and commands, autoharness distills and maintains skills automatically.
Full comparison → SkillOptMicrosoft's text-space optimizer that trains a frozen agent's skill document like weights — rollouts, bounded edits, validation-gated updates — and ships a compact best_skill.md.
Why switchBoth evolve agent skills from real sessions; autoharness learns, merges and prunes continuously inside Claude Code with no eval, SkillOpt (incl. its nightly Sleep mode) accepts a skill edit only when a held-out validation score improves.
Full comparison → OpenSpaceHKUDS's skill lifecycle layer for agents: retrieve the right skill per task, evaluate which ones actually work from real outcomes, share across agents and teammates, evolve with every run.
Why switchBoth run a skill lifecycle judged by real outcomes rather than benchmarks; openspace adds cross-agent sharing and retrieval, autoharness stays a local Claude Code plugin.
Full comparison → hivemindActiveloop's shared brain for agent TEAMS: traces from Claude Code, Codex, Cursor & co become reusable skills every teammate's agent can execute — cloud-backed, 25% cheaper on LoCoMo.
Why switchBoth turn agent traces into reusable skills; hivemind shares them across a team's agents via the cloud, autoharness keeps a self-maintaining local library for one user.
Full comparison →