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 skills/mirkosertic/mcpluceneserver/architecturenpx skills add mirkosertic/MCPLuceneServer --skill architecturegit clone --depth 1 https://github.com/mirkosertic/MCPLuceneServerWrote 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/mirkosertic/mcpluceneserver/architecture)<a href="https://agentmods.dev/skills/mirkosertic/mcpluceneserver/architecture"><img src="https://agentmods.dev/badge/skills/mirkosertic/mcpluceneserver/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 5.1 | $0.00036 | $0.01564 |
| Opus 5 | $0.00018 | $0.00782 |
| Sonnet 5 | $0.00007 | $0.00313 |
| Haiku 4.5 | $0.00004 | $0.00156 |
Grade A, and why
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 6d 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 — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Architecture Deep Dive
System Overview
┌─────────────────────────────────────────────────────┐
│ Claude Desktop / MCP Client │
└────────────────┬────────────────────────────────────┘
│ STDIO Transport (JSON-RPC)
▼
┌─────────────────────────────────────────────────────┐
│ MCP Java SDK (STDIO Transport) │
│ ┌──────────────────────────────────────────────┐ │
│ │ LuceneSearchTools (MCP Tools) │ │
│ └──────────────────────────────────────────────┘ │
└─────────┬───────────────────────────┬───────────────┘
│ │
▼ ▼
┌──────────────────────┐ ┌──────────────────────┐
│ LuceneIndexService │ │ DocumentCrawler │
│ - Search & Index │ │ Service │
│ - NRT Manager │ │ - File Discovery │
│ - Admin Operations │ │ - Content Extraction │
└──────────┬───────────┘ └──────────┬───────────┘
│ │
▼ ▼
┌─────────────────────────────────────────────────────┐
│ Apache Lucene 10.3 + Apache Tika │
└─────────────────────────────────────────────────────┘
Design Decisions
Why Plain Java (no Spring)?
- Fast startup (~1 second) - critical for MCP subprocess
- Smaller JAR (~45MB)
- Direct control over lifecycle
- Trade-off: Manual dependency wiring in
LuceneserverApplication.java
Why STDIO Transport?
- Required for Claude Desktop integration
- No network security concerns
- Consequence: Console logging MUST be disabled in production (
deployedprofile)
Why Multi-Analyzer Pipeline?
The server uses multiple analyzer chains optimized for different search patterns. Each document is indexed with several shadow fields, each using a different analyzer:
UnicodeNormalizingAnalyzer: Primary content field with NFKC normalization, diacritic folding, ligature expansionReverseUnicodeNormalizingAnalyzer: Reversed tokens for efficient leading wildcard queries (*vertrag)OpenNLPLemmatizingAnalyzer: Dictionary-based lemmatization for German and English (handles irregular forms)GermanTransliteratingAnalyzer: Maps ASCII digraphs to umlauts (Mueller→Müller)
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.
- 6d ago First seen · 152 lines · 36 tokens per session scan A 828dca1781c0
architecture is a skill published in the GitHub repository mirkosertic/MCPLuceneServer (5 stars, last pushed yesterday), licensed Apache-2.0. It adds 36 tokens to every session and 1,564 once invoked, about $0.0002 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
hevy-cli
Manage Hevy workouts from the terminal: read and summarize training data, search exercise templates, and create or update workouts, routines, exercises, folders, or body measurements.
help-flow
Help about Rosetta: explains capabilities and usage.
external-lib-flow
Workflow for onboarding an external private library so AI can use it without source access.
memora
Use when working with persistent memory across sessions, storing/retrieving knowledge, managing TODOs/issues, or when context from previous sessions would be helpful.
coding-flow
Light coding workflow: features, fixes, refactors, unit tests, etc.; scales small to large.
adhoc-flow
Workflow for the rest of tasks: lightweight documentation, build, track, synchronize, etc.