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 geoffsdesk/memanto-mcp --skill memanto-memorygit clone --depth 1 https://github.com/geoffsdesk/memanto-mcpWrote 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/geoffsdesk/memanto-mcp/memanto-memory)<a href="https://agentmods.dev/skills/geoffsdesk/memanto-mcp/memanto-memory"><img src="https://agentmods.dev/badge/skills/geoffsdesk/memanto-mcp/memanto-memory/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/geoffsdesk/memanto-mcp/memanto-memory"><img src="https://agentmods.dev/badge/skills/geoffsdesk/memanto-mcp/memanto-memory.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00137 | $0.02634 |
| Opus 5 | $0.00068 | $0.01317 |
| Sonnet 5 | $0.00027 | $0.00527 |
| Haiku 4.5 | $0.00014 | $0.00263 |
Grade A, and why
memanto-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 9d 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 — 246 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Memanto memory (enterprise build)
This plugin gives Claude typed, persistent memory that survives across
sessions and can be safely used inside multi-tenant B2B agents. Memory is
organized into agents (named scopes), each containing typed records
(fact, decision, commitment, …) with confidence and provenance.
The skill assumes a regulated-or-sensitive deployment: legal, finance, healthcare, customer ops at scale, internal knowledge work where the audit trail and tenant boundary are first-class. Adapt freely for less sensitive use, but keep the discipline.
The scoping rule (read this first)
Every memory must belong to exactly one tenant boundary. Cross-tenant recall is the highest-impact failure mode in production memory systems — "no results found" is the wrong response when the boundary is wrong; a loud refusal is the right one.
This server uses agent_id as the scoping primitive. Treat it as a
composite key:
agent_id = "<tenant>:<workspace>:<actor>"
For example: acme:matter-4711:user-jdoe, globex:deal-q3-fy26:team-legal,
tenant-7:account-mega:agent-onboarding.
Rules:
- Never write a memory whose
agent_idis missing the tenant component. If the user hasn't named one, ask before writing — do not default to a user-named agent in a multi-tenant context. - For finer axes (jurisdiction, region, product line, conflict-of-interest
group), use
tags. Tags are filterable on recall and are the right place for cross-cutting metadata. - Before any
memanto_recall*call, verify theagent_idyou are about to query is the one the current request is authorized for. If you are unsure, callmemanto_list_agentsand confirm with the user. - When a recall returns memories whose tags or content suggest a different tenant than the active one, stop and surface the mismatch. Do not silently include them.
Choose the right verb
There are three reads, and the choice matters more than people expect.
memanto_recall— semantic search returning raw hits. Use this when you (Claude) will compose the response. This is the default for enterprise work, especially anything client-facing.memanto_recall_current— same, but excludes superseded / expired / deleted memories. Use when "what is the current position" matters more than "what have we ever stored".memanto_recall_as_of— bi-temporal recall as of an ISO date. Use when the user asks "what did we know on ", "what was our position before ", or you are reconstructing a decision rationale for an audit.memanto_answer— RAG-synthesized answer from memanto's built-in LLM. Use only when the user explicitly wants memanto's synthesis or when comparing with your own answer.
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.
- 9d ago First seen · 246 lines · 137 tokens per session scan A 7ccf31259213
memanto-memory is a skill published in the GitHub repository geoffsdesk/memanto-mcp (3 stars, last pushed 4mo ago), licensed MIT. It adds 137 tokens to every session and 2,634 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
memanto-companion
Inspect and manage the cross-session engineering memory that Memanto maintains for your Claude Code skills. Use when the user asks what Memanto remembers, wants to see their engineering profile, manually recall context for a skill, or store a decision. The automatic lifecycle hooks handle capture/injection on their…
weekly-digests
Generate a serial week-by-week narrative digest of a project's full claude-mem timeline. Splits the timeline into per-ISO-week files, then runs one consecutive subagent per week — each receiving the prior week's carry-forward block — to produce one chapter per ISO week of data. Use when asked for "weekly digests"…
cloud-sync
Set up or check claude-mem cloud sync with cmem.ai Pro. Use when the user says "set up cloud sync", "sync my memories", "cmem pro", "cloud backup", "sync status", or wants their memory database backed up or synced to their cmem.ai account.
long-context
Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques. Use when processing long documents (32k-128k+ tokens), extending pre-trained models beyond original context limits, or implementing efficient positional encodings. Covers rotary embeddings, attention biases…
hive.note-taking
Maintain a free-form scratchpad of decisions, extracted values, and open questions so context pruning doesn't lose anything you still need.
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…