From a Personal Hermes Memory Plugin to a Skill, MCP Server, and Official Listings in Just 8 Days
I originally just wanted to switch to a long-term memory solution for the Hermes Agent to address the issue of running out of free quota on Mem0 SaaS. But the project kept expanding, and in the end, I ended up developing a Hermes plugin, a standalone runtime, an Agent Skill, an MCP server, and publishing it all on PyPI. The plugin was even listed in the official Hermes directory, and the MCP made…
The user initially aimed to switch Hermes Agent to a long-term memory solution on Mem0 SaaS to avoid running out of free quota. However, the project expanded and ended up creating a Hermes plugin, standalone runtime, an Agent Skill, an MCP server, and listing it in official directories.
The project was documented in a chronological account, starting from October 6 to October 13. The main project repository, https://github.com/cnkang/hermes-plugin-qdrant-memory, was built to migrate memories from Mem0 SaaS to Qdrant. The user chose Ollama’s qwen3-embedding:4b as the local embedding model and Qdrant for storage, as it supports local, self-hosted, and cloud deployment methods.
During development, the user faced challenges such as pagination, rate limits, and HTTP 429 errors while exporting from Mem0. They addressed issues like identity verification, memory scope isolation, and permission boundaries for multi-user sessions. The user also focused on persistence and exception recovery, implementing SQLite WAL Ledger to record pending events, write operations, and execution status.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
