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 agentmods add skills/luabagg/agent-skills/memory-palacenpx skills add luabagg/agent-skills --skill memory-palacegit clone --depth 1 https://github.com/luabagg/agent-skillsWrote 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/luabagg/agent-skills/memory-palace)<a href="https://agentmods.dev/skills/luabagg/agent-skills/memory-palace"><img src="https://agentmods.dev/badge/skills/luabagg/agent-skills/memory-palace.svg" alt="Measured on agentmods" 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 | $0.00023 | $0.01683 |
| Opus 5 | $0.00012 | $0.00842 |
| Sonnet 5 | $0.00005 | $0.00337 |
| Haiku 4.5 | $0.00002 | $0.00168 |
Grade A, and why
memory-palace 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 yesterday.
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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Memory palace
Before touching the vault, resolve $VAULT using this precedence:
- Explicit path in the current user request, if provided.
MEMORY_PALACE_VAULT, if available in the runtime environment.- Persisted default at
~/.agents/memory-palace/config.json(vaultPath). - Current-directory fallback: walk upward from cwd and use the first directory that looks like this vault.
The persisted config is created with:
agent-skills config memory-palace --vault <path>
For WSL, prefer a WSL-accessible path (/mnt/c/...). The setup script converts Windows drive paths like C:\Users\... to /mnt/c/Users/... when running under WSL, validates the converted path, and saves the WSL path.
Treat the resolved path as $VAULT below. If no candidate resolves, stop and ask the user to run setup or provide an explicit vault path.
The vault follows the Karpathy LLM-wiki pattern. Raw sources go in raw/. Compiled knowledge lives under wiki/ as plain markdown with wikilinks. Daily notes live in journal/. Ongoing responsibilities live in areas/. Plain markdown only, so any markdown editor works.
You are the librarian. The user is the curator. They decide what enters the vault. You file it, link it, and keep it consistent.
When to invoke
Trigger this skill on explicit user intent:
- Ingest -- "file this into my vault", "add this to the wiki", "ingest this URL/file", "save this article into my notes", "remember this".
- Query -- "what does my vault say about X", "look up X in my notes", "do I have notes on Y", "answer from my vault".
- Lint -- "audit my vault", "check the wiki", "find duplicates / orphans / broken links".
Do not trigger when:
- The user is doing unrelated coding/writing and merely mentions a person, tool, or concept.
- The user asks a general question that doesn't reference their vault.
- You're inside a different project's session and the user hasn't asked you to touch the vault.
Vault layout (canonical)
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.
- yesterday Changed 9c1809dec3f8
- 5d ago First seen · 123 lines · 23 tokens per session scan A e13932c19430
memory-palace is a skill published in the GitHub repository luabagg/agent-skills (2 stars, last pushed yesterday), licensed MIT. It adds 23 tokens to every session and 1,683 once invoked, about $0.0001 per session on Opus 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
checkpoint
Write a committed handoff checkpoint so a fresh session can resume with zero loss. Use when crossing the 40% context line, on every stage transition, and at session end. Writes .trace/checkpoints/ - .md while the agent is still sharp.
context-engineering
Optimizes agent context setup. Use when starting a new session, when agent output quality degrades, when switching between tasks, or when you need to configure rules files and context for a project.
last30Days
Resolve "last30Days" to a concrete ISO date range relative to your run time — a rolling 30-day window ending today. Returns inclusive civil dates plus exact UTC instants so you have temporal context without computing dates by hand. Read-only: no writes, no network. Use before a "last 30 days" / trailing-month task…
thisQuarter
Resolve "thisQuarter" to a concrete ISO date range relative to your run time — this quarter so far (quarter start → today). Returns inclusive civil dates plus exact UTC instants so you have temporal context without computing dates by hand. Read-only: no writes, no network. Use before a quarter-to-date task (QTD…
comet-memory
使用于 Comet 需要根据有界语义评审包判断是否值得保存个人记忆、候选、更新、遗忘或跳过时。.
md-fetch-summarize
Fetch a URL and return a concise markdown summary of its content. Read-only: no files are written; the summary is returned as output only. Use when asked to "fetch and summarize", "summarize this URL", "what does this page say", or "get the content of ". Proactively suggest when the user pastes a URL and asks what it…