Dev Open-Sources AI Codebase Memory Agent With Evidence-Backed Recall
haricharareddy · reddit · 2026-09-30
A developer released AI Codebase Memory Agent, an open-source project that builds a persistent knowledge layer for repositories: ingestion → knowledge extraction → memory → retrieval → evidence-backed answers. Key design: retrieved info stays linked to its source/evidence rather than trusting the LLM's answer as ground truth. Current features include GitHub repo ingestion, knowledge extraction, persistent memory, and evidence tracking. Open questions include keeping memory useful as the codebase evolves.
More from coding & agent
- One prompt to architect a full Jev decision layer with Claude, author claims — iamrobotbear · 2026-09-30
- ChatGPT subscription now works inside 60+ partner products including Devin, OpenCode and Lovable — TheMoonMidas · 2026-09-30
- OpenAI DevDay gives AI agents their own computer — from chatting to assigning work — 141_1337 · 2026-09-30
- Agent wired to macro datasets from the 1930s watches the economy 24/7 — virattt · 2026-09-30
- Making games with Opus 5.5? You need to be Pinterestmaxxing first — nptacek · 2026-09-30
- Conductor adds Sign in with ChatGPT to bring Codex subscriptions over — charlieholtz · 2026-09-30