Browse / Software development / Agents-Memory
Agents-Memory
haochencheng
Shared Engineering Brain
Open haochencheng/Agents-Memory → On the boundary of the four rules.
Works with any agent
ships AGENTS.md, the cross-tool instruction file.
In its own words
English
Agents-Memory is a Shared Engineering Brain for AI coding agents. It gives teams a reusable runtime for engineering memory, standards, planning bundles, and validation gates across real repositories.
System Architecture
Five layers inside the Brain:
Knowledge Layer (Wiki) Architecture
The Memory layer includes a structured wiki knowledge graph. This is a zoom-in on the Memory layer, not a separate system.
Wiki page anatomy (compiled_truth + append-only timeline):
Trust Signals
- CI is public and mirrors the local install, compile, test, and docs-check path.
- CI is split into
testsanddocsjobs for clearer branch protection. - CHANGELOG.md and docs/ops/release-checklist.md govern public release execution.
- SECURITY.md, SUPPORT.md, and CONTRIBUTING.md define collaboration paths up front.
Documentation
Full documentation map: docs/README.md
Open-Source Boundary
Public repository content is limited to code, templates, standards, profiles, and docs. The following are local runtime artifacts and should not be committed:
The repository uses public examples under templates/ so real project context, private paths, and runtime data do not leak into the open-source tree.
Contributing
Before shipping behavior changes, update code, matching docs, and matching tests or validation scripts together.
Quoting 50% of the README. Read the rest on GitHub ↗
Read from the repository's git tree on 2026-08-17.