Persistent memory for Claude Code - powered by Graphiti knowledge graphs
SaferSkills independently audited lisa (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Primary interface for project memory, tasks, and knowledge. Routes natural language requests to appropriate capabilities.
Use when the user addresses "lisa" directly:
1) Parse user intent from "lisa" request 2) Route to appropriate underlying command:
lisa memory load --cachelisa memory load --cache --query "<topic>"lisa memory add "<text>" --cachelisa tasks list --cachelisa tasks add "<text>" --cachelisa storage status --cachelisa storage switch <mode> --cachelisa compile-skills3) Summarize results conversationally
When user asks for a retrospective, follow these steps:
1) Gather Changes: Run git diff HEAD~20 --stat or git log --oneline -20 to see recent changes 2) Analyze Patterns: Review the changes and identify:
3) Format Findings: Create a concise summary covering:
4) Save to Memory: Use memory add command to save findings:
lisa memory add "RETROSPECTIVE: <findings>" --cache5) Report: Summarize what was learned and saved
| User Says | Intent | Route To |
|---|---|---|
| "show memories", "recent memories", "what's stored" | recall | memory load |
| "what do you know about X", "recall X", "search X" | search | memory load --query X |
| "remember that X", "save this", "note that X" | remember | memory add X |
| "tasks", "what are we working on", "todo" | list tasks | tasks list |
| "add task X", "new task X", "create task X" | add task | tasks add X |
| "what storage", "current mode", "storage status" | storage status | storage status |
| "switch to local", "use docker", "local mode" | switch local | storage switch local |
| "switch to zep", "use cloud", "zep-cloud mode" | switch zep-cloud | storage switch zep-cloud |
| "do a retrospective", "retrospective on session" | retrospective | git diff + memory add |
| "what did we learn", "session learnings" | retrospective | git diff + memory add |
| "compile skills", "rebuild skills" | compile skills | compile-skills |
| "merge skill extensions", "apply local skills" | compile skills | compile-skills |
👧 lisa-> (emoji, space, "lisa >>", space)👧 lisa-> Recent Memories: for memory listings👧 lisa-> Tasks: for task listings👧 lisa-> for conversational responses👧 lisa-> Tasks: (neo4j) or 👧 lisa-> Recent Memories: (zep-cloud) 👧 lisa-> Tasks: (neo4j)
Active (3):
1. 📋 Newest task here (Jan 23)
2. 📋 Second newest (Jan 22)
3. 📋 Third newest (Jan 20)
Completed (2):
1. ✅ Recently completed (Jan 22)
2. ✅ Older completed (Jan 15)created_at descending (newest first) 👧 lisa-> Recent Memories: (neo4j)
1. **Memory title** (date)
- Details hereCLI commands return JSON:
{ status: "ok", action: "load", facts: [...] }{ status: "ok", action: "add", text: "..." }{ status: "ok", action: "list", tasks: [...] }{ status: "ok", action: "add", task: {...} }{ status: "ok", action: "status", mode: "local|zep-cloud", isConnected: true|false }{ status: "ok", action: "switch", previousMode: "...", newMode: "...", verified: true|false }{ status: "ok", action: "compile-skills", skillsDir: "...", results: [...], merged: N, skipped: N, errors: N }~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.