Experimenting with graph-based semantic memory for AI agents
The article discusses Graft, a local-first agentic memory system designed for AI coding agents. It aims to help these agents retain knowledge across sessions, preventing the loss of important insights and decisions. Graft is not a vector database but a tool for persistent reasoning that enhances the productivity of AI agents by allowing them to build on their past experiences.
- ▪Graft provides a persistent memory for AI coding agents that survives session resets and machine switches.
- ▪It captures decisions, gotchas, and root causes as agents solve problems, allowing for reusable knowledge.
- ▪The installation process is straightforward, requiring only a single binary and an SQLite file.
2 outlets in our directory ran this story, first to last over 26 hours. All of the coverage we found sits in one bucket: centre. That one-sidedness is itself worth noticing.
Hacker News (AI / LLM) files mainly under ai. We currently carry 3,312 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | GitHub |
| Canonical URL | https://github.com/AEndrix03/Graft |
| Publication time | Thu, 21 May 2026 00:36:26 +0000 |
| Retrieval time | 2026-05-21T00:45:03.143Z |
| Last seen | 2026-05-21T00:45:03.143Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | 68t8gR2cBDpo · 2 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
graft Local-first agentic memory for AI coding agents. Stop solving the same problems twice. Give Claude Code, Codex and any other agent a persistent memory that survives sessions, context resets and machine switches — locally, with no cloud and no API key. C11 · SQLite + sqlite-vec + FTS5 · llama.cpp + BGE-M3 · MessagePack · AF_UNIX socket · optional REST + 3D viewer Made for Claude Code · Codex · ChatGPT · Claude Desktop · Gemini CLI · Open Code · and your own microservices. Why Graft? AI coding agents are productive — but they forget everything when the session ends.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.