Search, save, retrieve, list, and forget memories via a Python script.
npx skills add https://github.com/patanet7/corvus --skill memory-patanet7
Please help me install this Agent Skill. Skill: memory Source: https://github.com/patanet7/corvus/tree/main/config/skills/tools/memory Command: npx skills add https://github.com/patanet7/corvus --skill memory-patanet7
This Skill allows agents to persist and retrieve contextual information across conversations, preventing repetitive questions and maintaining continuity.
memory search to quickly retrieve the user's last stated preference or a critical decision made in a previous session.Use the memory skill to search for memories related to the user's last request.
High-intent search queries and answers about installing and using this skill.
To persist agent context across sessions, you use a memory skill to save and retrieve information. This enables agents to recall past decisions and user preferences without asking repetitive questions.
The best way to retrieve previous user preferences is to perform a memory search at the start of a new conversation. You can query past records by text, domain, or record ID to quickly restore context.
Yes, you can filter saved memories by domain and visibility. When saving or searching for records, the tool provides specific options to set importance levels and visibility for contextual queries.
You can delete or forget specific agent memories by executing the forget action via the Python script. This allows you to remove outdated records and manage the persistence scope of your agent context.
No, this memory tool requires no external dependencies to manage context. It operates via a standalone Python script executing specific actions like search, save, and get for persistent records.
wormholecowboyScaffold MCP servers in TypeScript or Python with stdio or HTTP transports.
gonzalezpazmonicaRecord, store, and replay agent session activities with JSONL storage and markdown export.
winsorllcStore, recall, and forget key-value memories across sessions using JSON.
math-incStore and query embeddings with metadata filtering in a local vector database.