by 64envy64 · MCP Server · ★ 35
TraceBase Memory layer for coding agents. Your agents stop re-solving the same problem. Agents are stateless. They forget everything between sessions, so the same bug gets re-derived from scratch, the same file gets re-read, the same loop spins again — every run, on your tokens. TraceBase keeps the resolved work and feeds it back into the next run. What it catches Five failure modes agents hit at runtime — one runtime, five arms: Catches doom-loops mid-run on a six-turn window and suggests a redirect. Never overrides agent judgem
| Stars | 35 |
| Forks | 2 |
| Language | TypeScript |
| Category | MCP Server |
| License | MIT |
| Quality Score | 66.3453854873615/100 |
| Open Issues | 1 |
| Last Updated | 2026-05-28 |
| Created | 2026-04-06 |
| Platforms | claude-code, mcp, node |
| Est. Tokens | ~847k |
These tools work well together with tracebase for enhanced workflows:
Explore other popular mcp server tools:
tracebase is The runtime memory layer for AI agents. Atomic writes, in-narrative time, first-class deletion - so agent intelligence compounds over next runs.. It is categorized as a MCP Server with 35 GitHub stars.
tracebase is primarily written in TypeScript. It covers topics such as ai, ai-agents, chatgpt.
You can find installation instructions and usage details in the tracebase GitHub repository at github.com/64envy64/tracebase. The project has 35 stars and 2 forks, indicating an active community.
tracebase is released under the MIT license, making it free to use and modify according to the license terms.