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 DavidOsipov/nplg-dspace-mcp --skill georgian-newspaper-visual-analysisgit clone --depth 1 https://github.com/DavidOsipov/nplg-dspace-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/davidosipov/nplg-dspace-mcp/georgian-newspaper-visual-analysis)<a href="https://agentmods.dev/skills/davidosipov/nplg-dspace-mcp/georgian-newspaper-visual-analysis"><img src="https://agentmods.dev/badge/skills/davidosipov/nplg-dspace-mcp/georgian-newspaper-visual-analysis/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/davidosipov/nplg-dspace-mcp/georgian-newspaper-visual-analysis"><img src="https://agentmods.dev/badge/skills/davidosipov/nplg-dspace-mcp/georgian-newspaper-visual-analysis.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 | $0.00039 | $0.01402 |
| Opus 5 | $0.00019 | $0.00701 |
| Sonnet 5 | $0.00008 | $0.00280 |
| Haiku 4.5 | $0.00004 | $0.00140 |
Grade A, and why
georgian-newspaper-visual-analysis 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 12d 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 — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Georgian Newspaper Visual Analysis
Core rule
OCR is not authoritative for historical Georgian print. Treat repository metadata and the downloaded source PDF as provenance, then verify wording visually. Do not automatically run OCR or present OCR output as a quotation.
This is an instruction-only, client-executed workflow. It does not grant network, filesystem, installation, or execution authority. Follow the user's instructions and the client host's security policy throughout. This workflow requires MCP resource support. Not every MCP host will discover, load, or follow it automatically.
Retrieve one public PDF
- Call
search_documentsand select one canonical NPLG handle. - Call
get_document_metadatawith that handle. Preserve the title, date, creators, rights, collections, canonical handle, and metadata source. If the record isrestricted, stop PDF retrieval and report public metadata only. - Call
list_document_fileswith the same canonical handle. Select only an entry whoseaccess_statusis exactlypublicand whose reported format or filename plausibly indicates a PDF. Reported format and filename are hints, not proof. If no eligible entry exists, stop at metadata. - Retrieve only the exact returned
source_urlfor the selected bitstream ID. Never construct, alter, substitute, or reuse a previously saved URL. Apply the client host's egress allowlist, require HTTPS, send no ambient credentials or cookies, bound response bytes and elapsed time, limit redirects, and reject redirects outside the permitted origin. Store the response in client-controlled local temporary storage. - Verify locally that the response is a PDF using both its file signature and a safe PDF parser; an HTTP content type is not proof. Compute and preserve a local SHA-256 over the exact downloaded bytes before analysis.
If the client cannot enforce a bounded download, verify the PDF safely, or view the pages locally, report the workflow as blocked and stop at metadata. Do not install software, expand permissions, or send the PDF to another service without explicit user approval.
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.
- 12d ago First seen · 119 lines · 39 tokens per session scan A dbd7f1df2f41
georgian-newspaper-visual-analysis is a skill published in the GitHub repository DavidOsipov/nplg-dspace-mcp (0 stars, last pushed 5d ago), licensed MIT. It adds 39 tokens to every session and 1,402 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
pptx
Create polished, editable PowerPoint files for slide decks, presentations, and explicit .pptx requests.
xlsx
Create polished Excel workbooks for budgets, trackers, tables, and explicit .xlsx spreadsheet requests.
docx
Create polished, editable Microsoft Word documents such as reports, letters, proposals, and handbooks.
Create polished PDF files for requests such as PDFs, resumes, CVs, reports, letters, one-pagers, and printable documents.
ai-strategy-report
Generate comprehensive AI strategy reports (15+ slides) in PowerPoint format. Analyzes company's digital readiness, identifies high-value AI use cases, creates implementation roadmaps, and provides ROI projections. Designed for consulting firm quality deliverables.
Convert PDF to Markdown using opendataloader-pdf, extracting text, tables, headings, and images with correct reading order.