43 plugins in our Memory & Context shelf (Things that help an agent remember.), ordered by what we can actually measure: our editor's picks first, then GitHub stars (checked 2026-10-26). Archived repos go to the bottom.
Stars measure popularity, not quality or safety. We haven't installed or audited these; every description is the project's own. Read the code before you install anything. Paid placements never affect this order.
Most-starred right now: deja by vshulcz.
Search the coding sessions already on your disk — Codex, Claude Code, Cursor, opencode and nineteen more — including work from before it was installed, indexed locally with no model in the loop.
★ 1,308 · by vshulcz#2Automatically capture, compress, and inject session context back into future Codex sessions.
★ 14 · by 2kDarki#3DevOps layer for coding agents with flow, feedback, and memory that compounds between sessions.
★ 450 · by boshu2#4Mines selected evidence from local coding-agent sessions into private work, design, and writing profiles for Codex, Claude Code, and GitHub Copilot.
★ 299 · by ohad6k#5Query traces, debug exceptions, analyze sessions, and manage prompts via MCP tools.
★ 112 · by avivsinai#6Spec-driven development and context engineering for Claude Code, Cursor, Codex, and GitHub Copilot — backed by project context in Git.
★ 72 · by archcore-ai#7Hands the steps of a Codex session that produce no text — is the build done, which of these elements do I click, is this shell command safe — to Jev, TypeSafe AI's judgment model, with a PreToolUse command gate and typed escalation back to the model for anything it cannot decide.
★ 65 · by shitianfang#8Persistent cross-session memory for Codex powered by Honcho — lifecycle hooks capture each session and inject relevant context back at session start, so Codex remembers your preferences, projects, and decisions across restarts.
★ 55 · by plastic-labs#9Session orchestration for Claude Code, Codex, and Cursor IDE — structured planning, wave-based execution, VCS integration (GitLab + GitHub), quality gates, and clean session close-out with issue tracking.
★ 52 · by Kanevry#10Lightweight documentation memory for AI coding agents that scaffolds a 7-category docs tree, runs readiness checks, validates docs before finishing, and keeps a monthly changelog across Claude Code and Codex.
★ 44 · by MedAdemBHA#11Persistent graph-backed conversational memory for Codex that recalls project decisions, constraints, preferences, and outcomes across sessions.
★ 44 · by Abhigyan-Shekhar#12Ranks repo context for Codex with likely files, skill recommendations, agent rules, commands, warnings, and compact task-focused packs before editing.
★ 25 · by vishal2612200#13Long-term memory for Codex with no LLM in the loop: a local MCP server stores messages word for word in SQLite and recalls the ones that answer a question within a token budget.
★ 22 · by Thanatos9404#14Local-first MCP server for private project docs with hybrid BM25+vector search, tree-sitter code indexing, and automated linting for team-wide documentation standards.
★ 17 · by epicsagas#15MCP server exposing reasoning, code, anti-deception, and memory harness tools for Codex.
★ 16 · by ejentum#16Shared session bus MCP + skill so Codex can list, read, hand off, send, and spawn with Claude, Cursor, Grok, and friends without pasting context.
★ 15 · by abnegate#17Hook-enforced guardrails for Codex and Claude Code with blocked destructive commands, one-time human tokens for irreversible operations, context-blind verification agents, and SQLite agent memory recalled at session start.
★ 14 · by ariaxhan#18Knowledge-backed SolidWorks design and validation workflows for Codex with standards lookup, CAD evidence gates, and consent-based session learning.
★ 14 · by Erfouni#19Harness blueprint skill for turning vague agent ideas into concrete designs for request assembly, control loops, memory, permissions, recovery, and extension planes.
★ 13 · by dadwadw233#20Cross-session error gates that detect recurring tool-call failures and escalate remind to block, with one shared store across Codex, Claude Code, OpenCode, Gemini CLI, Cursor and more (MIT).
★ 13 · by WhiteBite#21Generate compact first-pass repository briefings for coding agents before deeper exploration.
★ 12 · by Rothschildiuk#22Strips praise, recaps and filler from agent answers while keeping every fact you act on — findings carry location and one-line fix, refusals carry the safe procedure, tutorials stay long-form — benchmarked on 2,479 preregistered live calls, 266 of them real Claude Code sessions, with all raw records committed.
★ 11 · by Maksim-Burtsev#23Controlled Upwork job search, qualification, and proposal submission sessions through a dedicated Chrome profile.
★ 7 · by klajdikkolaj#24Reconstruct evidence-backed developer work from authorized Codex and Claude history plus read-only Git, then turn it into daily standups, weekly reviews, sprint recaps, and memory reports.
★ 6 · by jquinteiroo#25One MCP for context management: bind self-improving playbooks, MCP tools, and API keys to the session.
★ 4 · by not-so-fat#26Preserves authoritative requirements and verification evidence across long-running Codex tasks and context compaction.
★ 4 · by GreenLv#27A persistent per-session task list for Codex CLI with a live, scrollable panel across terminals through native integrations or a launcher with tmux fallback.
★ 3 · by martinille#28Durable memory and shared graph state for Codex and OpenClaw agents, with live ValkyrAI schema awareness.
★ 3 · by ValkyrLabs#29Guided learning with Socratic practice, Feynman teach-back, spaced recall, evidence-based project assessment, and optional local Anki export.
★ 2 · by everclear077#30Save 71% on MCP tool call tokens by wrapping any server with GCF encoding, with session stats hook and setup skill.
★ 2 · by blackwell-systems#31Give Codex lasting project context in an Obsidian vault you control.
★ 2 · by DuncanMain#32Decode VINs, license plates, market value, vehicle history, recalls, liens, OBD codes, and more via the CarsXE API.
★ 1 · by carsxe#33Hosted librarian for agent Markdown memory via host MCP.
★ 1 · by zeyu-j#34Visually validate real local terminal UIs in Codex's in-app browser with screenshots and session evidence.
★ 1 · by bnc4vk#35Owner-local memory for Codex that recalls user and project evidence before each answer and carries source-attributed project context across sessions and supported agent tools.
★ 1 · by reshuibuduo#36Git-backed repository memory for Codex and other coding agents via MCP, with durable local knowledge for decisions, conventions, gotchas, and playbooks.
★ 1 · by MiguelMedeiros#37AI operations plugin for Codex that connects production telemetry, problem lifecycle context, and remediation workflows through Aient's MCP server.
★ 0 · by aient-ai#38Free recovery and continuity plugin for AI agents: resume prior sessions, capture state, process failures into a recovery plan, and remember across sessions through a hosted MCP server (works in Codex, Claude Code, Cursor, and VS Code).
★ 0 · by davidmosiah#39Hosted AI agent memory that learns from outcomes, over remote MCP with OAuth sign-in: tell it a recalled memory helped or misled and it re-ranks what comes back next, with shared rooms for multi-agent work.
★ 0 · by mnemoverse#40Checks every "done" claim in local Codex and Claude Code session logs against the same session's commands and prints the counts plus a shareable card, read-only and offline.
★ 0 · by Kloudedge-apex#41Agentic newsroom over a hosted MCP server with narrated briefings, a news-versus-prediction-market Divergence Index, and a searchable declassified archive.
★ 0 · by droplinked#42Local-first Codex work journal that automatically archives sessions for private search, audit, usage analytics, and long-term memory continuity without recurring model-token cost.
★ 0 · by HuaHuiJian#43Local-first memory lifecycle guidance for Codex agents with recall, evidence-backed lessons, privacy-safe memory capture, audit, consolidation, and explicit forgetting.
★ 0 · by TerminallyLazy