$ loading_
Manage token-efficient memory for AI coding agents with dedup, merge, and decay.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Leptin" yet — see the docs or source repo.
Use Leptin to manage this AI coding agent's long-term memory: deduplicate repeated debugging notes, merge similar implementation learnings, decay outdated information, and report token savings for this run.
A cleaned memory state with retained, merged, decayed items, plus token savings statistics.
Store key information from the last 30 code changes into Leptin memory, automatically merge repeated conventions, weaken obsolete decisions, and preserve the most important architectural context for future agent runs.
A more compact project memory store that future coding agents can reuse.
Analyze Leptin's savings ledger, explain which memory items save the most context tokens, identify content that should be removed due to decay, and summarize whether the current memory strategy is effective.
A memory cost-benefit analysis with optimization recommendations and removable items.
Persistent memory for coding agents across sessions
Give AI coding agents searchable local project memory with safe structured updates.
Give AI coding agents persistent local memory of your workflow and preferences.
Give AI coding agents persistent memory across sessions for decisions and debugging context.
Give AI coding agents persistent semantic memory with embedding-based retrieval.
Give AI coding agents persistent local shared memory across agents.