whereismyai
Goatedcontext
Goatedcontext: Local developer memory shared across repositories and coding agents using a local SQLite file.
Goatedcontext stores local developer memory and shares it between coding agents and across repositories. It runs entirely locally using a SQLite file, with no server, embeddings, vector database, or cloud sync required. It maintains preferences, proposals, and signals separately to steer agents without overriding explicit instructions or hard constraints. The tool integrates natively with Claude Code, Codex, and Cursor, and supports any agent capable of speaking MCP, invoking a subprocess, or reading AGENTS.md. It is designed for developers who frequently switch between repositories and want their coding preferences remembered automatically.
Goatedcontext at a glance
Key features
- Stores structured developer preferences locally in a SQLite file
- Shares context across repositories and between different coding agents
- Separates preferences, proposals, and decision signals
- Integrates natively with Claude Code, Codex, and Cursor
- Supports custom agents via MCP server, CLI, or AGENTS.md
- Provides narrow permission rules for safe agent memory writes
Use it for
- Remembering coding style and tool preferences across multiple repositories
- Sharing technology stack choices between different AI coding agents
- Managing durable developer instructions without manual command entry
Alternatives to Goatedcontext
- claude-mem — It shares persistent, searchable agent memory across sessions and devices.
- CodeGraphContext — Turns codebases into interactive knowledge graphs for AI assistants.
- ContextMCP — Maintains fresh, accurate context for AI agents using AST-aware chunking.
- Personal Context Manager — Manages global and project-specific coding rules in a Git repository.
- Context7 — Provides up-to-date documentation for AI agents and code editors.
- Context Mode — Saves up to 98 percent of context per session by keeping raw data out of the LLM.
- Cognicore — Persists agent context across tools, sessions, and terminals.
- jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
- Lcc — Cuts token counts significantly without losing exact session details.
- Claude Context — It reduces token costs and provides deep context for large codebases.
- Engram — Single Go binary with zero dependencies and built-in full-text search.
- Elephant-Goldfish Model — Uses deep-context agents to draft specs and zero-context subagents to review them.
All Goatedcontext alternatives