Search and retrieve relevant content from soul, user, and work markdown files.
npx skills add https://github.com/shane9coy/katana-agent --skill recall-shane9coy
Please help me install this Agent Skill. Skill: recall Source: https://github.com/shane9coy/katana-agent/tree/main/agent/skills/memory/recall Command: npx skills add https://github.com/shane9coy/katana-agent --skill recall-shane9coy
This Skill allows the AI to search and retrieve relevant information from its memory stores, enabling it to answer questions about past work, user preferences, and project status.
soul.md, user.md, and work.md for relevant information based on user queries.work.md for entries from the previous day and summarize the findings.Ask the AI to recall what you worked on yesterday.
High-intent search queries and answers about installing and using this skill.
Contextual memory retrieval works by parsing specific markdown files like `soul.md`, `user.md`, and `work.md` to find relevant information based on natural language queries. It synthesizes the retrieved content conversationally instead of dumping raw data.
To search past project details and user preferences, you can ask a natural language question. The system searches your `work.md` and `user.md` files to retrieve and summarize relevant memories regarding project status and preferences.
You need filesystem access to read and parse specific markdown files for relevant content. The retrieval process depends on accessing these files to search and synthesize memories from your local storage.
The best way to recall agent identity and past work is querying the system with natural language. It searches `soul.md` for identity and `work.md` for history, then presents the retrieved memories in a synthesized conversational manner.
Natural language retrieval does not support raw data dumps. It applies a conversational synthesis to the retrieved memories from your files, ensuring that the information is presented naturally rather than as raw markdown content.
GACLoveOrganize AI context and memory through structured markdown files.
nicograefAnalyze session transcripts or git history to extract actionable insights and friction points.
eterdisScan six PESTEL categories and update a watch list with alerts.
QredenceIngest markdown files into ChromaDB for semantic procedural memory retrieval.