gbrain-maintain
1273Skill Claude CodeCodex
Run periodic maintenance over the knowledge graph.
12,529 mods in this category, of every kind an agent can take. Each one carries what it costs per session, what the scan found, and whether it is the original.
Skill Claude CodeCodex
Run periodic maintenance over the knowledge graph.
Skill Claude Code
Persistent session memory and wallet for AI agents — works on Bedrock, Vertex, API, and Foundry where Anthropic's built-in memory doesn't. Auto-captures git commits, publishes, and file writes without manual remember() calls. Apache 2.0 licensed, self-hostable, MCP-native. 15 MCP tools: remember, recall, forget…
Skill Claude Code
Local content + seen-ledger for the social-fetch family. Stores every fetched URL (and any URL the agent records via Claude WebFetch / research tools) in a SQLite + FTS5 store + a markdown mirror tree. Use to answer "have we seen this URL?" / "what did we save about X?" / "store this WebFetch output for next time" …
MCP server Claude CodeCodexCursor +2
Open-source agent-memory MCP server. recall.works. Runs locally from the ai-recallworks Python package.
Plugin Claude Code
Bundles 2 skills, 3 hooks · 116 tokens together
Records the corrections you make to your Claude Code agent and (opt-in) enforces them, so the same mistake doesn't come back. Safe observe-only by default.
Plugin Claude Code
Bundles 2 skills, 3 commands, 1 agent · 186 tokens together
Fully automatic per-project cognitive memory. Auto-loads git-aware context at session start, auto-saves session summaries, auto-recalls on errors, and auto-consolidates every 10 sessions. 9 MCP tools for episodic, semantic, procedural, working, and pattern memories with hybrid FTS5 + vector search.
Plugin Claude Code
Persistent memory for Claude Code. Indexes your conversation transcripts and provides hybrid search (BM25 + vectors + reranker).
MCP server Claude CodeCodexCursor +2
Hosted ambient memory for AI agents. Retrieve context and save notes via MCP tools with OAuth. Remote server at mcp.kortexio.io.
Plugin Claude Code
A collection of agent skills backed by logbooks — structured, queryable, session-persistent working surfaces. Examples: design and create a logbook, run a multi-angle code review with per-PR findings logbook.
Command Claude Code
NEVER workspace/[name].md (wrong) ALWAYS workspace/company/[name].md (correct).
LuisSambrano/antigravity-config
Skill Claude CodeCodex
Sistema universal para crear y mantener notebooks de NotebookLM como Memoria Viva de Proyecto. Úsalo cuando el usuario describe una idea de proyecto, pide crear un "libro", registrar decisiones de arquitectura/diseño o auditar la consistencia del proyecto. Actúa como el cerebro central que cualquier IA puede leer…
g1itchbot8888-del/agent-memory
Skill Claude CodeCodex
Complete memory system for OpenClaw agents. Combines behavioral protocol (when to save) + auto-capture (heartbeat-enforced) + semantic search (recall) + maintenance (consolidation). Use for persistent memory, context recovery, answering "what did we discuss about X", and surviving context compaction. Includes…
MCP server Claude CodeCodexCursor +2
A portable markdown memory MCP server for any agent. Store durable, searchable memories as git-friendly markdown files. Runs locally from the smriti-mcp Python package.
maxbaluev/accreted-intelligence
Cursor rule Cursor
acc — the accreted scored memory (substrate contract).
Plugin Claude Code
Self-correcting memory for Claude Code. Facts you correct stay corrected, deletions leave a signed receipt you can verify offline, and storing something never calls a model. Also ships the checks: erasureresidue asks whether the bytes actually went — on any store, not just this one.
MCP server Claude CodeCodexCursor +2
Lossless context compression for LLMs: 2x-8x fewer tokens, byte-exact reconstruction, sha256-verified. Runs locally from the densely Python package.
Plugin Claude Code
Bundles 3 skills, 7 commands, 1 agent · 504 tokens together
Continuous-collaboration layer for Claude Code: living project model, outcome telemetry, event digest, and calibrated confidence. Fully local, install-once, applies to every project.
Plugin Claude Code
Bundles 11 skills, 5 hooks · 551 tokens together
Knowledge-mining engine: captures hard-won knowledge as you work, gates it through machine and human review, and shares it as installable knowledge plugins.
Plugin Claude Code
Bundles 6 skills, 2 hooks · 170 tokens together
Your agent's memory, in a repo you own: the tome CLI, skills (capture, query, ingest, retrospect, write-a-plan, pickup-task), a SessionStart hook that points every session at the vault, a scoped Stop sync reminder, a no-build browse frontend (tome serve), and tome init to scaffold a fresh vault.
MCP server Claude CodeCodexCursor +2
MCP server that turns an Obsidian vault into semantic memory for any coding agent. Runs locally from the second-brain-mcp Python package.
MCP server Claude CodeCodexCursor +2
Cross-project memory layer for AI coding agents — MCP server that maintains context, decisions, and knowledge across workspaces. Runs locally from the hive-memory npm package.
MCP server Claude CodeCodexCursor +2
MCP server wrapping pgmnemo ingest and recall for AI agent memory. Runs locally from the pgmnemo-mcp Python package.
Skill Claude CodeCodex
Lay material into TheBrain as a connected graph rather than a pile of notes. Use when asked to "put this article in my brain", "save this", "add it to the knowledge base", "break this down and record it" — about an article, a document, a discussion, a research result or the user's own thinking. Covers: how finely to…
RasputinKaiser/Self-Improvement-Plugin
Plugin Claude Code
Bundles 11 skills, 12 commands, 5 agents, 7 hooks, 1 MCP server · 830 tokens together
SIPS graph runtime and home-base for bounded task execution, memory fabric recall, verification, delegation, and lifecycle automation.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: