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.
git clone --depth 1 https://github.com/wuisabel-gif/MemWhaleInstalling this writes the configuration, not the program: `mw-mcp` has to be on your PATH already, or the server will not start.
Wrote 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/mcp/wuisabel-gif/memwhale/claude-desktop-memorywhale)<a href="https://agentmods.dev/mcp/wuisabel-gif/memwhale/claude-desktop-memorywhale"><img src="https://agentmods.dev/badge/mcp/wuisabel-gif/memwhale/claude-desktop-memorywhale/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/mcp/wuisabel-gif/memwhale/claude-desktop-memorywhale"><img src="https://agentmods.dev/badge/mcp/wuisabel-gif/memwhale/claude-desktop-memorywhale.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
memorywhale 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 1mo 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
{
"memorywhale": {
"command": "mw-mcp"
}
}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.
- 1mo ago First seen · 5 lines scan A 86d82e3e9688
memorywhale is an MCP server published in the GitHub repository wuisabel-gif/MemWhale (147 stars, last pushed today), licensed MIT. Its token cost is not measured: an MCP server costs its tool schemas, not its config file. 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 mcp servers, from other repositories
engram
Your local, private memory layer. Index your notes and files, then recall anything with citations and temporal reasoning — 100% on your machine. Runs locally from the @jnmetacode/engram npm package.
errorlens
Structured error recovery for MCP agents. Runs locally from the mcp-errorlens npm package.
memorylane
Agent-native memory layer — persistent context, semantic recall, session history, SDK, CLI, API, and MCP for AI agents and developers. Runs locally from the @talocode/memorylane npm package.
vectorlane
Open-source local vector memory engine for AI agents — ingest documents, chunk content, generate embeddings, search locally, return metadata/citations. Runs locally from the @talocode/vectorlane npm package.
datacron
Local-first MCP server that exposes a Markdown vault through MCP. Runs locally from the datacron Python package. Needs 1 environment variable to run.
fimemory
Local-first shared memory for AI tools. Your files, encrypted at rest by default; budgeted reads. Runs locally from the fimemory npm package. Needs 1 environment variable to run.