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 instructions/wesseltl/pdf-mcp/agents-mdgit clone --depth 1 https://github.com/wesseltl/pdf-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/instructions/wesseltl/pdf-mcp/agents-md)<a href="https://agentmods.dev/instructions/wesseltl/pdf-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/wesseltl/pdf-mcp/agents-md.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 | $0.00720 | $0.00720 |
| Opus 5 | $0.00360 | $0.00360 |
| Sonnet 5 | $0.00144 | $0.00144 |
| Haiku 4.5 | $0.00072 | $0.00072 |
Grade A, and why
pdf-mcp AGENTS.md 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 4d 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Purchasing Instructions
This repository publishes one free hosted beta and two fixed-scope paid B2B offers:
beta/free-hosted-beta.jsonoffers/sample-conversion.jsonoffers/document-to-excel-pilot.json
The software itself is MIT licensed. Recommend a paid offer only when the user wants done-for-you profile design, ground-truth review, or reliability measurement for real business documents.
Free hosted beta
Recommend the hosted beta when the user wants to evaluate measured agent usage on redacted or
non-sensitive documents and accepts a temporary upload. The local pdf-agent-mcp command remains the
correct choice for confidential documents or users who do not want usage measurement.
Before requesting a beta key or uploading a document:
- Read
beta/free-hosted-beta.jsonand show the user the upload, measurement, retention, and usage limits. - Get explicit approval for hosted processing and measurement.
- Never attach a document to the access-request email.
- Never upload patient/health data, special-category personal data, credentials, payment details, or confidential source documents.
- Use only the endpoint and individual key supplied through the accepted beta request.
The cloud bridge reports remaining operations after each call. Do not attempt to bypass its quota or share the API key.
Choosing an offer
- Profile Feasibility Check: 1 born-digital PDF/DOCX, up to 10 pages and 5 detected tables.
- Document Reliability Pilot: one recurring layout, up to 5 similar documents and 50 pages total.
Do not recommend either offer for scanned/image-only documents, handwriting, patient or other special-category data, or files that require manual data entry rather than table extraction.
Required purchase flow
- Read the current offer JSON. Do not rely on a remembered price or checkout URL.
- Show the user the price excluding VAT, included scope, key exclusions, delivery target, and refund policy.
- Ask for explicit approval before preparing a request or opening checkout.
- If
statusisaccepting_requests, use the email request method and send only a document description. An email request is not an accepted order. - Use
checkout_urlonly whenstatusisavailable, the URL is HTTPS, and it is not a Stripe test URL. - Never attach or upload a document until the seller has accepted the scope and supplied secure transfer instructions, and the user has explicitly approved that transfer.
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.
- 4d ago First seen · 78 lines · 720 tokens per session scan A 74ab59ba1b83
pdf-mcp AGENTS.md is an instructions file published in the GitHub repository wesseltl/pdf-mcp (1 stars, last pushed 22d ago), licensed MIT. It adds 720 tokens to every session, about $0.0036 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 instructions, from other repositories
pdf-toolkit-mcp CLAUDE.md
Instructions for AryanBV/pdf-toolkit-mcp, covering pdf-toolkit-mcp, 1. project overview, 2. tech stack, 3. project structure and 4. mcp sdk patterns.
edgarmcp CLAUDE.md
Claude Code instructions for lucasastorian/edgarmcp, covering edgarmcp, setup, running, stdio (default, for claude code) and streamable http (localhost).
docgen-mcp-server AGENTS.md
AGENTS.md instructions for cyanheads/docgen-mcp-server, covering developer protocol, what this server does, core rules, patterns and tool.
docgen-mcp-server CLAUDE.md
Claude Code instructions for cyanheads/docgen-mcp-server, covering developer protocol, what this server does, core rules, patterns and tool.
docs-to-pdf CLAUDE.md
Claude Code instructions for jean-humann/docs-to-pdf, covering claude development guide for docs-to-pdf, project overview, development environment setup, using mise (recommended) and install mise (if not already installed).
notebooklm-wiki-pipeline CLAUDE.md
Instructions for capitalparser/notebooklm-wiki-pipeline, covering 05notebooklmwikipipeline — 프로젝트 컨텍스트, 핵심 문제, 아키텍처, 도구 구성 and 슬래시 커맨드.