Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/RobThePCGuy/Claude-Patent-Creatornpx agentmods add skills/robthepcguy/claude-patent-creator/index-managerWrote 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/robthepcguy/claude-patent-creator/index-manager)<a href="https://agentmods.dev/skills/robthepcguy/claude-patent-creator/index-manager"><img src="https://agentmods.dev/badge/skills/robthepcguy/claude-patent-creator/index-manager/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/robthepcguy/claude-patent-creator/index-manager"><img src="https://agentmods.dev/badge/skills/robthepcguy/claude-patent-creator/index-manager.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 | $0.00020 | $0.01095 |
| Opus 5 | $0.00010 | $0.00548 |
| Sonnet 5 | $0.00004 | $0.00219 |
| Haiku 4.5 | $0.00002 | $0.00110 |
Grade C, and why
index-manager scanned grade C with 2 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 10d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf mcp_server/index_backup_* Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
wget https://www.uspto.gov/web/offices/pac/mpep/mpep-2900.pdf -O pdfs/mpep-2900.pdf Copies of this mod
1 near-identical copy found in the catalogue:
- index-manager — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Index Manager Skill
Expert system for managing MPEP search index lifecycle: PDF downloads, index building, maintenance, updates, optimization.
FOR CLAUDE: All dependencies installed, system operational.
- Go directly to appropriate phase
- Scripts/tools in mcp_server/
- Use patent-creator CLI when available
- Only run diagnostics if operations fail
When to Use
Building/rebuilding MPEP index, corruption/missing files, optimization, adding content, troubleshooting.
Index Lifecycle
PDFs Not Present -> Download (2-5 min, 500MB)
-> Extract & Parse (500MB data)
-> Generate Embeddings (5-10 min GPU, 35-65 min CPU)
-> Build FAISS + BM25 Indexes
-> Index Ready (mcp_server/index/)
-> Maintenance (Verify -> Optimize -> Update)
Phase 1: PDF Management
Check Status:
ls pdfs/ # Should show mpep-*.pdf, consolidated_laws.pdf, consolidated_rules.pdf
Download PDFs:
patent-creator download-mpep
# Or: python install.py (Select "Download MPEP PDFs")
Verify Integrity:
python -c "
import fitz
from pathlib import Path
for pdf in Path('pdfs').glob('*.pdf'):
try:
doc = fitz.open(pdf)
print(f'[OK] {pdf.name}: {len(doc)} pages')
doc.close()
except Exception as e:
print(f'[X] {pdf.name}: ERROR - {e}')
"
Phase 2: Index Building
patent-creator rebuild-index
# Or: python mcp_server/server.py --rebuild-index
Timeline:
- Load PDFs: 30s
- Extract text: 1-2 min
- Chunk text (500 tokens): 30s
- Generate embeddings: 5-10 min (GPU) or 35-65 min (CPU)
- Build FAISS/BM25: 1 min
- Save to disk: 10s
Total: 5-15 min (GPU) or 35-65 min (CPU)
Custom Build:
from mcp_server.mpep_search import MPEPIndex
index = MPEPIndex(use_hyde=False)
index.build_index(
chunk_size=500,
overlap=50,
batch_size=32 # Reduce to 16/8 if OOM
)
Phase 3: Verification
# Check files
ls -lh mcp_server/index/
# Expected: mpep_index.faiss (~150MB), mpep_metadata.json (~80MB), mpep_bm25.pkl (~60MB)
# Verify health
patent-creator health
# Should show: [OK] MPEP Index: Ready (12,543 chunks)
# Manual test
python -c "
from mcp_server.mpep_search import MPEPIndex
index = MPEPIndex()
print(f'Chunks: {len(index.chunks)}')
results = index.search('claim definiteness', top_k=3)
print(f'Search results: {len(results)}')
"
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.
- 10d ago First seen · 174 lines · 20 tokens per session scan C 7df015cd06c3
index-manager is a skill published in the GitHub repository RobThePCGuy/Claude-Patent-Creator (180 stars, last pushed yesterday), licensed MIT. It adds 20 tokens to every session and 1,095 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
plagiarism-checker
Scans lyrics for phrases that may match existing songs using web search and LLM knowledge. Use before release to check for unintentional borrowing.
compensation-and-leveling
Builds and maintains the leveling framework and pay structure — level definitions, salary bands, benchmarking, pay equity, and how raises and promotions are decided. Use this to design or revise leveling, set or adjust salary bands, benchmark against market, handle a compensation request or counteroffer, run a review…
employment-compliance
Covers the employment rules that carry real penalties — exempt and non-exempt classification, overtime and hours, employee versus contractor status, work authorization and recordkeeping, accommodation requests, and the notices and retention obligations that go with them. Use this to classify a role, review a…
sanity-best-practices
Sanity development best practices for schema design, GROQ queries, TypeGen, Visual Editing, images, Portable Text, Studio structure, localization, migrations, Sanity Functions, webhooks, Blueprints, and framework integrations such as Next.js, Nuxt, Astro, Remix, SvelteKit, Angular, Hydrogen, and the App SDK. Use this…
video-perception
Use when the user mentions a video file (.mp4, .mov, .avi, .mkv, .webm), a YouTube URL, asks to watch/analyze/review a video, or references video content in conversation.
tax-ebookkeeping-context
Background information about Japan’s Electronic Bookkeeping Act, which sets rules for storing accounting records, scanned documents, and data received through electronic transactions.