58b6626, the version that was reviewed and scanned here, and Claude Code installs exactly that commit. Newer commits reach this catalog only after a new review. License: MIT, by pdftomarkdown.dev.
- Skills
/pdftomarkdown:pdf-to-markdown
This is the author's README: its install commands may install another version. pdftomarkdown@aitmpl installs commit 58b6626, the one listed and scanned here.
pdftomarkdown — PDF reading for Claude Code
A Claude Code plugin that lets Claude read any PDF: scanned documents, complex tables, multi-column layouts, math, and Chinese/Japanese/Korean text. Under the hood it runs npx pdftomarkdown, which converts the PDF to markdown with a hosted vision-language OCR service at pdftomarkdown.dev.
Once installed, just ask Claude things like "summarize this PDF", "extract the table from invoice.pdf", or "what does this paper say about X?" — the skill activates automatically.
Install
In Claude Code:
/plugin marketplace add ThiloReintjes/pdftomarkdown-skill
/plugin install pdftomarkdown@pdftomarkdownRequirements: Node 18+ (for npx). No API key needed to start — the free demo tier converts the first page of any PDF. For full documents, get a free API key (20 starter pages once) in 30 seconds at https://pdftomarkdown.dev/auth/github and set:
export PDFTOMARKDOWN_API_KEY=your_key_hereManual install (without the plugin system)
Copy the skill into your personal skills directory:
mkdir -p ~/.claude/skills
cp -r skills/pdf-to-markdown ~/.claude/skills/What the skill teaches Claude
- Convert local files or URLs:
npx pdftomarkdown <file-or-url> - Allow up to 11 minutes for the synchronous request; conversion time varies with the document and provider capacity
- Save long documents to a file and read sections, instead of flooding context
- Bound cost with
--max-pages, relay actionable API error messages - Suggest the free API key when the demo tier's page-1 limit is hit
Links
- API docs: https://pdftomarkdown.dev/docs/
- OpenAPI spec: https://pdftomarkdown.dev/openapi.json
- npm CLI: https://www.npmjs.com/package/pdftomarkdown
License
MIT
pdftomarkdown — PDF reading for Claude Code
A Claude Code plugin that lets Claude read any PDF: scanned documents, complex tables, multi-column layouts, math, and Chinese/Japanese/Korean text. Under the hood it runs npx pdftomarkdown, which converts the PDF to markdown with a hosted vision-language OCR service at pdftomarkdown.dev.
Once installed, just ask Claude things like "summarize this PDF", "extract the table from invoice.pdf", or "what does this paper say about X?" — the skill activates automatically.
Install
In Claude Code:
/plugin marketplace add ThiloReintjes/pdftomarkdown-skill
/plugin install pdftomarkdown@pdftomarkdownRequirements: Node 18+ (for npx). No API key needed to start — the free demo tier converts the first page of any PDF. For full documents, get a free API key (20 starter pages once) in 30 seconds at https://pdftomarkdown.dev/auth/github and set:
export PDFTOMARKDOWN_API_KEY=your_key_hereManual install (without the plugin system)
Copy the skill into your personal skills directory:
mkdir -p ~/.claude/skills
cp -r skills/pdf-to-markdown ~/.claude/skills/What the skill teaches Claude
- Convert local files or URLs:
npx pdftomarkdown <file-or-url> - Allow up to 11 minutes for the synchronous request; conversion time varies with the document and provider capacity
- Save long documents to a file and read sections, instead of flooding context
- Bound cost with
--max-pages, relay actionable API error messages - Suggest the free API key when the demo tier's page-1 limit is hit
Links
- API docs: https://pdftomarkdown.dev/docs/
- OpenAPI spec: https://pdftomarkdown.dev/openapi.json
- npm CLI: https://www.npmjs.com/package/pdftomarkdown
License
MIT
Scanned by
NVIDIA SkillSpector
Automated static analysis (no LLM) · v2.12.0 · 2026-11-22
14 observations in 3 categories
- MCP Rug Pull 12
- Agent Snooping 1
- Rogue Agent 1
SkillSpector looks for patterns such as shell commands, credential access, network calls or unpinned dependencies. An observation is something worth reading in the files before you install, not a confirmed vulnerability. Components that run code locally, like hooks, MCP servers and security tooling, naturally show more of them. Open one to see the lines it points at.
Skill EnumerationAgent Snooping · README.md:27
| 25 | |
| 26 | ```sh |
| 27 | mkdir -p ~/.claude/skills |
| 28 | cp -r skills/pdf-to-markdown ~/.claude/skills/ |
| 29 | ``` |
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.MCP Rug Pull · .claude-plugin/marketplace.json:11
| 9 | "name": "pdftomarkdown", |
| 10 | "source": "./", |
| 11 | "description": "Lets Claude read any PDF — scanned documents, tables, multi-column layouts, CJK — by converting it to markdown with npx pdftomarkdown." |
| 12 | } |
| 13 | ] |
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.MCP Rug Pull · .claude-plugin/plugin.json:3
| 1 | { |
| 2 | "name": "pdftomarkdown", |
| 3 | "description": "Lets Claude read any PDF — scanned documents, tables, multi-column layouts, CJK — by converting it to markdown with npx pdftomarkdown.", |
| 4 | "version": "0.1.0", |
| 5 | "author": { |
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.MCP Rug Pull · SKILL.md:3
Code excerpt not available for this observation.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update. ×2MCP Rug Pull · README.md:3
| 1 | # pdftomarkdown — PDF reading for Claude Code |
| 2 | |
| 3 | A Claude Code plugin that lets Claude read any PDF: scanned documents, complex tables, multi-column layouts, math, and Chinese/Japanese/Korean text. Under the hood it runs [`npx pdftomarkdown`](https://www.npmjs.com/package/pdftomarkdown), |
| 4 | |
| 5 | Once installed, just ask Claude things like *"summarize this PDF"*, *"extract the table from invoice.pdf"*, or *"what does this paper say about X?"* — the skill activates automatically. |
Showing 6 of 14 observations. Learn more about NVIDIA SkillSpector or how components are reviewed.