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/danielrosehill/claude-user-memory-plugin/remember-user-factnpx skills add danielrosehill/Claude-User-Memory-Plugin --skill remember-user-factgit clone --depth 1 https://github.com/danielrosehill/Claude-User-Memory-PluginWrote 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/danielrosehill/claude-user-memory-plugin/remember-user-fact)<a href="https://agentmods.dev/skills/danielrosehill/claude-user-memory-plugin/remember-user-fact"><img src="https://agentmods.dev/badge/skills/danielrosehill/claude-user-memory-plugin/remember-user-fact.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 | $0.00103 | $0.00834 |
| Opus 5 | $0.00051 | $0.00417 |
| Sonnet 5 | $0.00021 | $0.00167 |
| Haiku 4.5 | $0.00010 | $0.00083 |
Grade A, and why
remember-user-fact 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 5d 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 — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Remember a user fact
When you learn something about the user that would be useful next session, save it now — don't defer to end-of-session. Small, atomic saves are easier to retrieve and update than one big dump later.
When to invoke
- Explicit request: "remember that…", "save this for next time", "store in memory".
- Correction: the user tells you to change your behavior ("don't do X", "always do Y", "stop assuming Z"). These are high-value — save them so the next session doesn't repeat the mistake.
- Confirmed preference: the user states a stable preference with reasoning, not just an offhand comment. Reasoning is the signal that it's durable.
- Durable project context: you learn a client name, ongoing initiative, long-running constraint, or recurring stakeholder that will matter beyond this session.
- Identity/role details: role changes, new responsibilities, location changes, tooling switches.
Do not invoke for:
- Transient conversation state (current file, current task).
- Information already in code, git, or project docs — memory is for facts you cannot derive from the filesystem.
- One-off frustrations or snap opinions the user may not hold tomorrow.
- Secrets, credentials, tokens, or anything sensitive.
How to run it
- Load the config — read
.claude/memory-config.mdin the workspace. That file names the backend, the exact MCP tool to call for "add", and the scope parameters (index/namespace/project_id/etc.) for the chosen context. If the file is missing, stop and ask the user to install one (copy the plugin'stemplates/memory-config.example.md). - Pick the context — apply the deduction rule from
CONTEXT.md. If unsure, ask one short question ("save this to personal or work memory?"). - Call the configured add tool with the context's scope parameters. Construct the record according to the schema in
memory-config.md; at minimum:- A self-contained natural-language statement of the fact, written from the user's perspective as if they were telling a future assistant. Example:
"I prefer pnpm over npm for new Node projects — npm's lockfile churn has burned me twice."Include the why if it was given; future-you uses the why to judge edge cases. - An absolute
YYYY-MM-DDdate. - Optional metadata the config recommends (type=preference/correction/project-context, topic tags, etc.).
- A self-contained natural-language statement of the fact, written from the user's perspective as if they were telling a future assistant. Example:
- Before saving, check for duplicates — run a quick semantic search for the same topic using the backend's search tool. If a near-identical memory already exists, prefer an update over creating a second record. Duplicates pollute retrieval.
- Confirm briefly — one short line: "Saved to personal memory." or "Saved to work memory." Do not quote the full memory back.
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.
- 5d ago First seen · 41 lines · 103 tokens per session scan A c5c037ed3883
remember-user-fact is a skill published in the GitHub repository danielrosehill/Claude-User-Memory-Plugin (2 stars, last pushed 4mo ago), licensed MIT. It adds 103 tokens to every session and 834 once invoked, about $0.0005 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
media-ingest
Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.
mem0-oss-to-platform
Plan and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…
agent-memory
../../../engineering/agent-memory/skills/agent-memory/SKILL.md.
memory
Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.
ha-data-stores
Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…