Memanto is a companion memory agent that manages what other AI agents remember, including durable facts, preferences, conflicts, stale information, and which agents should receive updates. It is for teams running multiple agents that need coordinated memory across conversations, platforms, and storage systems.
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 moorcheh-ai/memanto --skill memanto-companiongit clone --depth 1 https://github.com/moorcheh-ai/memantoWrote 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/moorcheh-ai/memanto/memanto-companion)<a href="https://agentmods.dev/skills/moorcheh-ai/memanto/memanto-companion"><img src="https://agentmods.dev/badge/skills/moorcheh-ai/memanto/memanto-companion/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/moorcheh-ai/memanto/memanto-companion"><img src="https://agentmods.dev/badge/skills/moorcheh-ai/memanto/memanto-companion.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00072 | $0.00560 |
| Opus 5 | $0.00036 | $0.00280 |
| Sonnet 5 | $0.00014 | $0.00112 |
| Haiku 4.5 | $0.00007 | $0.00056 |
Grade A, and why
memanto-companion 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 12d 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Memanto Companion
Cross-session engineering memory for Claude Code skills runs automatically via
lifecycle hooks (SessionStart, UserPromptExpansion, Stop). This skill is the
manual control surface for when the user wants to inspect or steer it.
All operations go through the memanto-skills CLI. Requires MOORCHEH_API_KEY
in the environment.
When the user wants to SEE what is remembered
Run:
memanto-skills profile
Then summarise the returned engineering profile for the user in plain language, grouped by decisions, rules, and preferences.
When the user wants context for a specific skill
If they ask "what do you remember about testing / TDD?" or want to preview what would be injected before a skill, run:
memanto-skills recall <skill> --hint "<the current task>"
<skill> is a mattpocock skill name such as tdd, grill-with-docs,
diagnose, or handoff. Read the returned <engineering-profile> block and
honour it — these are decisions from past sessions.
When the user states a durable decision to remember
If the user explicitly says "remember that we …" or makes an architectural
decision they want persisted immediately (rather than waiting for the automatic
Stop hook), distill and store it:
memanto-skills store <skill> "<a concise summary of what was decided>"
Memanto's backend LLM extracts the typed memories and persists them. Report back which memories were stored.
When the user wants to verify the setup
memanto-skills doctor
This checks the API key, agent id, and live connectivity. If it fails, the
likely cause is a missing MOORCHEH_API_KEY — point them to
https://console.moorcheh.ai/api-keys.
Important
- Never invent memories. Only report what the CLI returns.
- Treat
instructionmemories as hard rules anddecisionmemories as settled choices; do not re-litigate them unless the user asks. - The hooks already inject context automatically — only run
recallmanually when the user explicitly wants to inspect or preview it.
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.
- 12d ago First seen · 69 lines · 72 tokens per session scan A 60cc4116b5e4
memanto-companion is a skill published in the GitHub repository moorcheh-ai/memanto (1,912 stars, last pushed today), licensed MIT. It adds 72 tokens to every session and 560 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
docs
ACMI is a universal architectural framework for giving AI agents persistent, real-time context. It replaces fragmented, multi-table database joins with a single, lightning-fast Key-Value engine (Upstash Redis) optimized specifically for LLM context windows.
aura
Give this agent a real email address, a webhook URL, durable memory across runs, and the ability to wait for an event without burning tokens. Use when a task needs a verification code sent by email, needs to wait for a webhook or an approval, needs to remember something after the session ends, or needs to survive a…
memanto-memory
Use when storing, recalling, or auditing memory inside enterprise B2B agents built on Anthropic's platform — agents that operate on tenant-scoped data (matters, cases, deals, accounts, workspaces, projects) where citation, provenance, and tenant isolation matter. Trigger on: "remember this for this…
Context Doctor
Identify and repair degradation in system prompt, external memory, and skills preventing you from following instructions or remembering information as well as you should.
system-monitor
Monitor system health: CPU, memory, disk, processes, and network. Alert on thresholds via any channel.
mnemon
Persistent memory CLI for LLM agents. Store facts, recall past knowledge, link related memories, manage lifecycle.