# Agent memory · Tech watch

[Canonical source](<https://sebastien.rigaux.be/en/tech-watch/tags/agent-memory>)

Preserving and retrieving useful context across agent steps or sessions.

Related tags : [AI Agents](<https://sebastien.rigaux.be/en/tech-watch/tags/ai-agents>) · [Knowledge graphs](<https://sebastien.rigaux.be/en/tech-watch/tags/knowledge-graphs>) · [RAG](<https://sebastien.rigaux.be/en/tech-watch/tags/rag>)

## [oh-my-hermes: coordinating models, skills, and verification](<https://sebastien.rigaux.be/en/tech-watch/links/oh-my-hermes-coordinating-models-skills-and-verification>)

oh-my-hermes adds an orchestration layer to Hermes Agent: it maps work categories to models and reasoning effort, loads skills on demand, and can turn a goal into a graph of parallel tasks. Each task keeps a separate write scope, while a distinct node verifies the outcome.

A notable architectural choice is **separating preparation from execution evidence**: a plan is not treated as completed work, and an agent’s claim that it finished does not validate the result. The project also offers review-gated memory and a dedicated TUI, without replacing or patching Hermes Agent.

2026-10-28 · [Markdown](<https://sebastien.rigaux.be/en/tech-watch/links/oh-my-hermes-coordinating-models-skills-and-verification.md>) · [Original resource](<https://github.com/rlaope/oh-my-hermes>)

## [Engram: persistent memory for MCP-compatible AI agents](<https://sebastien.rigaux.be/en/tech-watch/links/engram-persistent-memory-for-mcp-compatible-ai-agents>)

Engram provides **persistent memory** for AI agents, compatible with any agent supporting the MCP protocol (Claude Code, Gemini CLI, Codex, VS Code, etc.). The tool runs as a single Go binary with no external dependencies, storing memories in a SQLite file with built-in FTS5 search. It features a TUI, Git sync, and a self-hostable cloud extension, while remaining local-first and open source under the MIT license.

2026-10-16 · [Markdown](<https://sebastien.rigaux.be/en/tech-watch/links/engram-persistent-memory-for-mcp-compatible-ai-agents.md>) · [Original resource](<https://engram.gentlemanprogramming.com/>)

## [Rowboat: an AI coworker with persistent context](<https://sebastien.rigaux.be/en/tech-watch/links/rowboat-an-ai-coworker-with-persistent-context>)

Rowboat presents itself as a collaborative assistant with memory and tool integrations. Its repository makes the architecture and integrations available for inspection beyond the product overview.

2026-02-15 · [Markdown](<https://sebastien.rigaux.be/en/tech-watch/links/rowboat-an-ai-coworker-with-persistent-context.md>) · [Original resource](<https://github.com/rowboatlabs/rowboat>)

## [Claude-Mem carries context across agent sessions](<https://sebastien.rigaux.be/en/tech-watch/links/claude-mem-carries-context-across-agent-sessions>)

Claude-Mem captures session activity, compresses it, and brings relevant details into later sessions. Its repository describes persistent memory for several coding agents.

2026-02-04 · [Markdown](<https://sebastien.rigaux.be/en/tech-watch/links/claude-mem-carries-context-across-agent-sessions.md>) · [Original resource](<https://github.com/thedotmack/claude-mem>)

## [Mastra 101: learn to build agents inside your editor](<https://sebastien.rigaux.be/en/tech-watch/links/mastra-101-learn-to-build-agents-inside-your-editor>)

This practical course guides the creation and deployment of Mastra agents. Tools, memory and MCP are introduced through a guided workflow inside the code editor.

2025-06-11 · [Markdown](<https://sebastien.rigaux.be/en/tech-watch/links/mastra-101-learn-to-build-agents-inside-your-editor.md>) · [Original resource](<https://mastra.ai/course>)