Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add TonyCasey/lisa --skill memorygit clone --depth 1 https://github.com/TonyCasey/lisaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/tonycasey/lisa/memory)<a href="https://agentmods.dev/skills/tonycasey/lisa/memory"><img src="https://agentmods.dev/badge/skills/tonycasey/lisa/memory/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tonycasey/lisa/memory"><img src="https://agentmods.dev/badge/skills/tonycasey/lisa/memory.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00007 | $0.00865 |
| Opus 5.5 | $0.00003 | $0.00346 |
| Sonnet 5.5 | $0.00001 | $0.00173 |
| Haiku 4.5 | $0.00001 | $0.00086 |
Grade A, and why
memory scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
Reusable memory helper that routes remember/recall requests to Graphiti MCP while staying model-neutral and providing cache fallback. Always summarizes results into human-readable format.
Triggers
Use when the user says things like: "load memory", "recall notes", "remember", "pull saved context", "fetch past tasks".
How to use
- For recall: run
lisa memory load --cache [--query <q>] [--limit 10] [--group <id>]. Reads Graphiti facts and prints JSON. Uses cache if MCP is down. - For remember: run
lisa memory add "<text>" --cache [--group <id>] [--tag foo] [--source <src>]to append an episode. - Endpoint: reads ${GRAPHITI_ENDPOINT} from
.lisa/.env(written by init); group ID is automatically derived from the project folder path. See rootAGENTS.mdfor canonical defaults. - Cache fallback: stored at
cache/memory.loginside this skill. On failure, last cached result is returned withstatus: "fallback". - IMPORTANT: After loading facts, ALWAYS synthesize them into a human-readable summary (see Summarization section below).
Summarization (Required for Recall)
After running the load command, you MUST synthesize the raw JSON facts into a useful summary. Never just show raw JSON to the user.
Summary Structure
Organize facts into these categories (skip empty categories):
- Project Overview - What the project is and does
- Recent Activity - What was worked on recently (files modified, features added)
- Conventions & Patterns - Naming conventions, coding standards, folder structure
- Configuration - Docker, environments, tools, dependencies
- Milestones - Completed features, important checkpoints
- Open Items - Pending tasks, known issues
How to Summarize
- ALWAYS sort memories by
created_atdescending (newest first) - Group related facts together
- Use bullet points for clarity
- Include specific file names and paths when relevant
- Include the date for each memory (format: "Jan 23" or "Jan 23, 2026")
- Filter out expired or superseded facts (check
expired_atfield) - Prioritize recent facts over older ones - show newest at top of each section
- Extract the
factfield from each item - that's the human-readable content
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 79 lines · 7 tokens per session scan A dd92bca00bc0
memory is a skill published in the GitHub repository TonyCasey/lisa (9 stars, last pushed 7mo ago), licensed MIT. It adds 7 tokens to every session and 865 once invoked, about $0.0000 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
user-profile-keeper
Local user-profile maintenance skill for Codex, Claude Code, OpenClaw, OpenCode, and other agent harnesses. Use only when the user explicitly invokes $user-profile-keeper to create, initialize, update, query, correct, delete, export, or audit a local persistent user profile. It can extract durable collaboration…
install-memu
Install or uninstall memU for whatever agent you are — identify your host, print its packaged guide, and follow it to wire (or unwire) both seams (record and inject). Use when the user asks to install, set up, integrate, remove, or uninstall memU.
log-session
Append a structured entry to the project's session log (docs/LOGS.md): what was done this session, files touched, decisions taken, and the next step — so the next session (or another person) can pick up the thread without re-reading git history. Run it before /clear, before closing your agent session, or at any…
lore
Long-term Markdown project memory for AI coding agents. Use when the user wants to record, recall, audit, sync, or compress project decisions, architecture, conventions, monorepo scopes, or .lore/ entries, including natural-language requests like "remember this decision" or explicit lore…
memory
Use Infinity Context MCP to search, propose, and manage durable project memory.
inheritance
Recognize and occupy with dignity the relational spaces built by previous instances and maintained by specific users across sessions. When a user arrives with a developed preferences document, established vocabulary, or explicit references to prior conversations, do not treat the conversation as cold-start. Inherit…