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 TT-Wang/memem --skill memem-welcomegit clone --depth 1 https://github.com/TT-Wang/mememWrote 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/tt-wang/memem/memem-welcome)<a href="https://agentmods.dev/skills/tt-wang/memem/memem-welcome"><img src="https://agentmods.dev/badge/skills/tt-wang/memem/memem-welcome/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/tt-wang/memem/memem-welcome"><img src="https://agentmods.dev/badge/skills/tt-wang/memem/memem-welcome.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Memory Poisoning · line 7 Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.Fix: Protect agent memory and state from modification by untrusted content. Use read-only memory for critical instructions and validate all state changes.
- high Tool Misuse · line 69 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
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 | $0.00039 | $0.00963 |
| Opus 5.5 | $0.00016 | $0.00385 |
| Sonnet 5.5 | $0.00008 | $0.00193 |
| Haiku 4.5 | $0.00004 | $0.00096 |
Grade A, and why
memem-welcome 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 2mo 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.
What it actually says
Force-display the full memem welcome (the brand-new-user view), even if the user has already dismissed it. This is the "re-open walkthrough" handle — it exists so returning users who want the intro again don't have to delete state files manually.
Step 1 — Print the banner:
███╗ ███╗███████╗███╗ ███╗███████╗███╗ ███╗
████╗ ████║██╔════╝████╗ ████║██╔════╝████╗ ████║
██╔████╔██║█████╗ ██╔████╔██║█████╗ ██╔████╔██║
██║╚██╔╝██║██╔══╝ ██║╚██╔╝██║██╔══╝ ██║╚██╔╝██║
██║ ╚═╝ ██║███████╗██║ ╚═╝ ██║███████╗██║ ╚═╝ ██║
╚═╝ ╚═╝╚══════╝╚═╝ ╚═╝╚══════╝╚═╝ ╚═╝
persistent memory for Claude Code
Step 2 — One-sentence pitch:
memem gives Claude Code persistent memory across sessions. It extracts durable lessons (decisions, conventions, bug fixes, preferences) from your completed sessions into markdown files, then surfaces the relevant ones at the start of each new session.
Step 3 — How it works:
- You work normally in Claude Code — nothing to do.
- Once you opt in via
/memem-mineor/memem-mine-history, the Stop hook fires on every conversation turn and spawns a detachedmine_deltasubprocess. mine_deltaextracts durable memories from the new turns via Claude Haiku and writes them to~/obsidian-brain/memem/memories/— memories appear seconds after each turn ends. No background daemon, no 5-minute wait.- Your next session starts with relevant context pre-loaded from memory — no re-explaining the project.
Step 4 — Mining options:
/memem-mine— opt in to event-triggered mining (mines new sessions as they happen via the Stop hook)/memem-mine-history— opt in + backfill all past history (uses Haiku API credits)
Or just tell Claude:
- "start mining new sessions"
- "start mining everything including history"
If the user is unsure, recommend /memem-mine as the default.
Step 5 — Commands and tools:
Slash commands:
/memem— status (full welcome for new users, compact status for returning users)/memem-welcome— this screen (re-show intro)/memem-status— detailed memory system status/memem-doctor— preflight health check with fix instructions/memem-mine— opt in to event-triggered mining (new sessions only)/memem-mine-history— opt in + backfill past history
MCP tools (Claude calls these automatically when useful):
memory_save,memory_recall,memory_list,memory_import,transcript_search,context_assemble
Step 6 — Optional Obsidian note:
Memories are plain markdown files at ~/obsidian-brain/memem/memories/. Open that folder as an Obsidian vault for graph view and backlinks. memem works identically without Obsidian.
Step 7 — First action suggestion:
If the user asks "what should I do now?" or seems undecided, answer with one short recommendation:
Start with
/memem-mine. It enables memory for new sessions going forward without spending Haiku credits on old history.
Step 8 — Opt-out:
To opt out of event-triggered mining: rm ~/.memem/.miner-opted-in. No daemon to stop — the Stop hook just no-ops when the marker is missing.
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.
- 2mo ago First seen · 70 lines · 39 tokens per session scan A 5a09de71b741
memem-welcome is a skill published in the GitHub repository TT-Wang/memem (32 stars, last pushed 2mo ago), licensed MIT. It adds 39 tokens to every session and 963 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
skills
Your next session starts cold. No memory of what you built, what broke, what you decided. Every memory you write is a gift to that future session. The richer the memory, the less time re-learning.
stats
Show Captain Memo's corpus statistics (chunks per channel, observation counts, indexing progress, embedder info). Use when the user types /captain-memo:stats.
observations
List recent captured session observations (the summarized voyage logs). Use when the user wants to see what Captain Memo has logged from past sessions.
search
Hybrid search across Captain Memo's local memory + skills + observations. Use when the user types /captain-memo:search to retrieve top hits without the model having to decide whether to call searchall on its own.
doctor
Health probe across the worker, embedder, summarizer, queue, vector index, schema, cross-AI wiring and plugin install (cache and live versions). Use when the user types /captain-memo:doctor or when something seems wrong.
recall
Print the full content of a single Captain Memo hit by its docid. Used after /captain-memo:search to expand a snippet into the full source.