Launched this week
Agentmemory
Persistent memory for Claude Code, Codex & coding agents
550 followers
Persistent memory for Claude Code, Codex & coding agents
550 followers
You can now give Hermes, Claude Code, and Codex infinite memory. Agentmemory is trending on GitHub with 5,000+ Stars. CLAUDE md dumps 22,000+ tokens into context at 240 observations agentmemory: 1,900 tokens. same observations. 92% less. At 1,000 observations, 80% of your built-in memories become invisible. agentmemory keeps 100% searchable. benchmarked on 240 real coding sessions → Up to 95% fewer tokens per session → 200x more tool calls before hitting context limits → 100% open source
Having trouble with large data sets. Just spent a few days stress-testing agentmemory on a real-world workload.
Dataset: ~674 Claude Code sessions / ~370K observations / ~108 hand-curated memories, imported from 1.5 GB of jsonl history across 24 projects.
Graph extracted to ~75K nodes / ~6K edges. All running locally with OpenRouter (Gemini 2.5 Flash Lite for LLM, OpenAI text-embedding-3-small for embeddings).
Bugs hit, all confirmed as open GitHub issues:
- #502 — Graph auto-trigger never fires on session end (the session.stopped topic has no publisher in shipped code)
- #544 — /memories and /export endpoints return 500 on large corpus
- #587/584 — Buffer pool corruption causing worker crash loops (filed today!)
- #474 — agentmemory stop leaves iii engine running with stale function registrations
- #538 — Viewer's "Rebuild Graph" button calls /graph/build which doesn't exist server-side
- #455 — No dimension override means switching embedding models requires AGENTMEMORY_DROP_STALE_INDEX=true (which I learned the hard way actually persisted on every restart in my hacked-together approach)
Workarounds I built: custom hourly LaunchAgent calling /agentmemory/graph/extract directly to compensate for #502; chunked batching (50 obs/request, file-based payload via --data-binary @file) for /graph/extract to avoid "Argument list too long"
Net result: Sessions API + Memories tab work fine, live capture works, recall context injection works at session start (~1300 tokens of project history). The graph viz endpoint times out at our size and the worker crash loop makes things unstable when querying graph-heavy endpoints. LaunchAgent workarounds just caused more chaos, so reverted to started terminal frontend launch/operations to simplify. But the issues remain.
Fortunately, even after wiping and starting over halfway through, the total OpenRouter spend across all this was only about ~$15. The concept and the architecture are solid; the scale-handling needs another iteration or two before it's production-stable for a real coding workflow. Watching the GitHub issues.
Great work, excited to see this puppy land with the kinks worked out! Pausing usage until its a bit further along.
Well done team! How do you detect when a stored memory contradicts current code state or is pruning still manual?
Agentmemory
Persistent memory across sessions is one of those things that sounds like a dev tool problem but actually changes how useful AI agents are in practice. Right now every session with Claude Code starts from scratch — re-explaining context, re-loading preferences. Curious how Agentmemory handles conflicts when the same context gets updated across sessions. Does it merge, overwrite, or flag it for review?
Agentmemory
Congrats on the launch.
2 questions:
Will this impact more usage on tokens? since the agent need looking around and search on newer chats?
Will the memory be persistent only in CLI agents or also on their desktop application as Codex, Claude, Cursor
Agentmemory
@rohit_ghumare that’s nice, what about 2#
Does it work only on CLIs?
92% token reduction is huge if it holds on real codebases. Curious how agentmemory handles conflicting observations: when newer context contradicts older stored memory, does recency win automatically or is there a manual override?
Cool project, how are you handling caching to ensure that it doesn't reprocess tokens unnecessarily in longer conversations?
Hey! Love it. How well would it help with handling pivots and knowing how my seed-stage startup's narrative/pitch deck and product spec changes over time? I've got canonical documents set up in Cursor, but it still takes a LOT of tidying work and any new scratch brainstorming files ruin the source of truth...