MCP servers and skills: search, filter by where they run, and compare.
Nothing matches every word exactly, so these are the closest matches.
Long-term memory for coding agents, kept as plain Markdown in your repository. Your agent records the decisions, rules, corrections and pitfalls of a project in .memory/, reads them back before it wor
SkillMITFew starsNo key
Correction-driven memory for coding agents: a GitHub template where each correction becomes a slim fact, a pytest test and a commit. No API key, no database, no service. For Claude Code and any agent
SkillMIT★ 2Few starsNo key
A portable memory layer that makes AI agents reliable across apps and sessions. Persistent, verifiable, and fully under your control.
SkillApache-2.0★ 131No key
Verified, token-aware project memory for Codex with selective loading, context budgets, and Git-reviewable Markdown.
SkillMIT★ 1Few starsNo key
Tend an AI agent's long-term memory: route what to keep, pick what to load per session, audit for rot, and forget for real. EN + 中文.
SkillMITFew starsNo key
A privacy-first, review-gated memory workflow for multiple AI assistants, devices, and terminal instances.
SkillMITFew starsNo key
Persistent, self-maintaining project memory for AI agents: a memory.md 'brain' that agents read and update automatically. Works with Claude Code, opencode (tested), and Codex; optional zero-config enf
SkillMITFew starsNo key
pi 原生永久记忆与自进化技能，名字即宣言：FACTS。F=Frugal 速进场(省token)｜A=Accurate 存取一致(零上下文读回不走样)｜C=Card 卡式(按场景建卡索引)｜T=Trusted 可信度分级(★✓~?)｜S=Self-evolving 任务卡自进化(坑表回写)。零依赖，ALWAYS ACTIVE。
SkillMITFew starsNo key
A fully local Markdown memory system for AI agents. No cloud, no lock-in.
SkillMIT★ 1Few starsNo key
Agent Skill: layer-by-layer Windows process memory forensics - topology, VMMap regions, minidump heap scans, disk correlation, evidence-grade attribution
SkillMITFew starsNo key
📦 CloudArc bounded-memory benchmark: streaming pack/unpack SLOs, multi-file .vibo dedup, remote-search telemetry, CI gates #agentskills #benchmark #memory #vibo
SkillLink onlyFew stars
Portable, auditable Obsidian-first memory continuity for Codex and other AI agents.
SkillMITFew starsNo key
OpenClaw GenPark Memory Vault - Cryptographically partitioned hierarchical memory vault with Ebbinghaus time-decay and HMAC namespace isolation.
SkillMIT★ 7No key
Instagram MCP server for Claude Code and AI agents. Posts, insights, comments, DMs, publishing, hashtag research and competitor analysis across three tiers of access.
SkillMIT★ 2Few starsNo key
🧠 Project memory (MEMORY.md) plus session handoffs (HANDOFF.md) so the next session picks up where this one stopped - Agent Skill for Claude Code, Cline, Cursor, Codex
SkillMITFew starsNo key
Recover current project decisions with sourced recall, outcome-tested lessons and read-only readiness reports that expose stale evidence and missing checkpoints.
SkillMITFew starsNo key
Agent skill that installs or upgrades dev-time project memory: a thin PROJECT.md plus lazy STATE, DECISIONS, JOURNAL and specs.
SkillMITFew starsNo key
OneBrain integration for OpenClaw — AI memory for autonomous agents
SkillMITFew starsNo key
Self-improving harness plugin for Antigravity: verified lessons, scoped AGENTS.md rules, reusable skills, and local cross-session memory.
SkillApache-2.0★ 1Few starsNo key
记忆导演：家庭照片与视频的本地记忆馆。Windows 主机、本地人物识别与回忆影片、Agent 部署 Skill，附 iPhone 上传客户端工程。
SkillMITFew starsNo key
Consent-driven, portable Personal Context protocol for AI agents — compile a minimal frozen Context View per task, return new insights as user-approved patches. Not an AI memory backend.
SkillMITFew starsNo key
帮助 AI Agent 在长期软件项目中记住背景、检查并修复任务结果，并能在不同工具之间接着做。
SkillMITFew starsNo key
Claude Code /handover skill — condense the current session into a self-contained handover prompt; paste into any fresh session to continue (session 交接 prompt 產生器)
SkillMITFew starsNo key
Durable project memory, authority and directory design, resumable handoffs and separately authorized Git closeout.
SkillApache-2.0★ 2Few starsNo key
Who stands behind it
