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/litopys-dev/litopys/litopys-memorynpx skills add litopys-dev/litopys --skill litopys-memorygit clone --depth 1 https://github.com/litopys-dev/litopysWrote 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/litopys-dev/litopys/litopys-memory)<a href="https://agentmods.dev/skills/litopys-dev/litopys/litopys-memory"><img src="https://agentmods.dev/badge/skills/litopys-dev/litopys/litopys-memory.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 | $0.00131 | $0.01620 |
| Opus 5 | $0.00066 | $0.00810 |
| Sonnet 5 | $0.00026 | $0.00324 |
| Haiku 4.5 | $0.00013 | $0.00162 |
Grade A, and why
litopys-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 4d 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 — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Litopys Memory
Persistent graph memory across sessions. The MCP server already injects 5 baseline rules (SEARCH FIRST · CREATE ON LEARNING · LINK AFTER CREATING · AVOID DUPLICATES · QUALITY THRESHOLD). This skill goes further — covers what those rules don't: graph traversal, temporal awareness, conflict resolution, constraint-aware writes.
Customize this skill
After installing, open SKILL.md and replace the generic trigger description above with the exact project and system names from your graph. Run litopys startup-context or ask litopys_search("projects") to see them. This makes the skill fire at the right moments instead of relying on generic phrasing.
Iron Law
No factual claim about user-specific things without searching the graph first.
Applies to: the user's projects, systems, infrastructure, people, past decisions, recurring problems.
Does NOT apply to: generic programming questions, library docs, syntax, code review unrelated to the user's stack.
Four Disciplines
1. Reading — Search + Traverse (top pain)
Plain litopys_search returns flat results. The graph carries meaning in edges. Every search must be followed by traversal when the answer needs context.
Step 1: litopys_search("keyword") → find candidate nodes
Step 2: litopys_get(id) → full body if needed
Step 3: litopys_related(id) → MANDATORY for any non-trivial answer
Step 4: Check supersedes chains → incoming supersedes = node is outdated, jump to the newer source
Step 5: Check until field → until < today = tombstoned, ignore
Red flag: answering after step 1 alone. If you didn't traverse, you're using the graph as a flat key-value store — exactly the failure mode this skill exists to prevent.
Details and examples → references/search-and-traverse.md
2. Writing — Decide before creating
Three mutually exclusive paths when new information arrives:
| Situation | Action |
|---|---|
| Genuinely new entity, no similar node found | litopys_create + litopys_link |
| Existing fact has changed (RAM upgraded, version bumped, role changed) | Create new node, litopys_link with supersedes from NEW to OLD ("new supersedes old") |
| Two facts disagree but both valid in different contexts | litopys_link with conflicts_with |
| Same thing already exists | Don't create — update the existing node body via dashboard or skip |
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 4d ago First seen · 154 lines · 131 tokens per session scan A 5e56fc8c08c4
litopys-memory is a skill published in the GitHub repository litopys-dev/litopys (3 stars, last pushed 1mo ago), licensed MIT. It adds 131 tokens to every session and 1,620 once invoked, about $0.0007 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
ori-memory
Persistent agent memory with learning retrieval. Knowledge graph on markdown files — capture insights, decisions, research, and learnings during work, then retrieve them weeks or months later. Use when knowledge is too valuable to lose but too much to inject into every prompt.
kb
Set up, evolve, or operate a hraness/kb local-first Markdown knowledge base for coding-agent memory. Use when a user asks to design KB conventions or a recurring KB ritual; search or query a KB or Obsidian vault; load or recover repository context, plans, decisions, concepts, backlinks, semantic search, or Git…
zettelkasten-brain
第二记忆大脑 — 自动保存知识到 Zettelkasten，建立双向链接， 评估知识健康度，需要时检索。.
personal-knowledge-graph
Use when maintaining a LoomKG/Obsidian knowledge graph.
mneme-prime
Use when the user starts a new task that may have prior vault context worth surfacing. Invokes the mnemeprime MCP tool to build a token-budgeted preamble of recent sessions and topic matches.
mneme-search
Use when the user asks a factual question whose answer might live in the vault. Invokes mnemesearch. Production mnemesearch is FTS5 BM25. The experimental feature-hashed lexical-vector backend is not wired into MCP. KG enrichment is gated to summarize or timeline when full-profile graph state is active.