prompt

Capture prompts as deduplicated memory episodes in Graphiti.

9|1|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/TonyCasey/lisa --skill prompt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prompt
Source: https://github.com/TonyCasey/lisa/tree/main/src/project/.lisa/skills/prompt
Command: npx skills add https://github.com/TonyCasey/lisa --skill prompt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Capture prompts into Graphiti memory (episodes) to enable rapid recall, auditing, and context retention across projects and AI sessions.

Core Features & Use Cases

  • Persist prompts as memory episodes with optional role and kind metadata for richer recall.
  • Deduplicate prompts using a fingerprint tag to avoid duplicates unless forced.
  • Enable quick recall of recent prompts during coding sessions or AI interactions across projects.

Quick Start

Use the prompt skill to save a new prompt as a memory episode for this repo/user. Example: lisa-prompt add --text "Describe the task" --role user --kind Observation --force

Frequently Asked Questions about prompt

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I save AI prompts as memory episodes for later recall?

To save AI prompts as memory episodes for recall, use the add command with your text and optional role or kind metadata, which logs the prompt into Graphiti memory for auditing across sessions.

How does fingerprint-based deduplication work for captured prompts?

Fingerprint-based deduplication prevents identical prompts from being saved as duplicate memory episodes by checking tags, but you can override this protection using the --force flag to ensure capture.

Do I need a Graphiti MCP endpoint to record prompt memory episodes?

Yes, recording prompt memory episodes requires a Graphiti MCP endpoint configured in your environment, defaulting to http://localhost:8010/mcp/ via the GRAPHITI_ENDPOINT variable.

Can I persist prompts across different coding sessions and projects?

You can persist prompts across coding sessions, chats, and tool integrations by capturing them as memory episodes, enabling rapid recall and context retention on a per-repo or per-user basis.

What is the best way to audit AI prompts used in past coding sessions?

The best way to audit AI prompts from past coding sessions is to capture them as memory episodes with metadata like role and kind, enabling quick recall and traceability within Graphiti memories.