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 rules/ralfhuesing/computetoai/mcp-server-architecturegit clone --depth 1 https://github.com/RalfHuesing/ComputeToAiWrote 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/rules/ralfhuesing/computetoai/mcp-server-architecture)<a href="https://agentmods.dev/rules/ralfhuesing/computetoai/mcp-server-architecture"><img src="https://agentmods.dev/badge/rules/ralfhuesing/computetoai/mcp-server-architecture.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 | $0.00000 | $0.00459 |
| Opus 5 | $0.00000 | $0.00230 |
| Sonnet 5 | $0.00000 | $0.00092 |
| Haiku 4.5 | $0.00000 | $0.00046 |
Grade A, and why
mcp-server-architecture 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 3d 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
MCP-Server-Architektur
Vollständige, verbindliche Fassung: Docs/02-Architektur-und-MCP.md. Kurzfassung:
- Transport: stdio (Standard-Transport des MCP-Python-SDK), kein Netzwerkport, keine Authentifizierung.
- Tool-Namenskonvention: MCP kennt keine verschachtelten Tool-Namensräume, die Hierarchie wird per Präfix abgebildet –
core_für Kern-Tools, Feature-Name als Präfix für Feature-Tools (z. B.finance_),calculations_für Berechnungen-Tools. - Selbstbeschreibung: Tools/Parameter beschreiben sich über ihr Pydantic-Schema selbst; zusätzlich stellt der Server
Docs/als MCP-Resources bereit – keine zweite, separat gepflegte Doku für den Agenten (sieheliving-documentation.mdc). - Settings-Datei: TOML, verweist u. a. auf das Arbeitsverzeichnis und das Logging-Level; Pfad standardmäßig plattformüblich, überschreibbar per Umgebungsvariable/Startargument. Generisches Beispiel unter
examples/. - Arbeitsverzeichnis: Wurzel mit einem Unterordner je Plan (kopierbare Pläne, siehe
Docs/00-Vision.md), nicht ein einzelner Plan-Ordner. Agent greift nie direkt auf diese Dateien zu, nur über Tools. - Logging: immer aktiv, gestuft – INFO ohne sensible Finanzwerte, DEBUG (nicht standardmäßig aktiv) mit vollständigen Tool-Parametern/-Ergebnissen. Ziel:
<Arbeitsverzeichnis>/logs/, plattformüblicher Fallback vor dem Laden der Settings-Datei. Python-Standardbibliothekloggingmit rotierenden Dateien, keine Zusatzpakete.
Bei Widerspruch zwischen dieser Kurzfassung und Docs/02 gilt Docs/02.
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.
- 3d ago First seen · 19 lines · 0 tokens per session scan A ffd9c0b3f0c6
mcp-server-architecture is a cursor rule published in the GitHub repository RalfHuesing/ComputeToAi (0 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 459 tokens. 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 cursor rules, from other repositories
cursorrules
When the user asks about lead scoring, use lead-scoring-ai-mcp tools: scorelead, addlead, updateleadactivity, getleadscore, getallleads.
cursorrules
When the user asks about web research, use web-research-mcp tools: websearch, browsepage, extractarticle, researchtopic, getweather.
cursorrules
When the user asks about nist rmf, use nist-rmf-ai-mcp tools: assessriskprofile, mapaiimpact, generateriskcontrols, crosswalktoeuaiact, createriskreport.
cursorrules
When the user asks about quiz generator, use quiz-generator-ai-mcp tools: generatequiz, validateanswers, generateflashcards, assessdifficulty.
cursorrules
When the user asks about social media, use social-media-ai-mcp tools: schedulepost, generatehashtags, analyzeengagement, plancontentcalendar, getaudienceinsights.
cursorrules
When the user asks about ecommerce, use ecommerce-ai-mcp tools: writeproductdescription, optimizepricing, summarizereviews, forecastinventory, generateseometa.