pdf-mcp AGENTS.md

pdf-mcp AGENTS.md is an instructions file for Codex, OpenCode from sandraschi/pdf-mcp. It costs 805 tokens per session, scanned A, original, MIT.

A set of project instructions for working on pdf-mcp, a service that provides PDF tools through an MCP server. It identifies the recommended reading order, important files, ports, and commands.

In plain words
What is it for?
It is for navigating the server, PDF processing tools, supporting services, web app, and development commands.
Why use it?
It helps an agent find the right parts of an unfamiliar PDF project before making changes.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/sandraschi/pdf-mcp/agents-md
Clone the repo
git clone --depth 1 https://github.com/sandraschi/pdf-mcp

Made for: Codex, OpenCode.

Wrote 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.

agentmods badge for pdf-mcp AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/sandraschi/pdf-mcp/agents-md.svg)](https://agentmods.dev/instructions/sandraschi/pdf-mcp/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/sandraschi/pdf-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/sandraschi/pdf-mcp/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 805 This file is loaded in full into every session.
When invoked 805 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00805 $0.00805
Opus 5 $0.00402 $0.00402
Sonnet 5 $0.00161 $0.00161
Haiku 4.5 $0.00081 $0.00081

Measured 3d ago against content hash b3d5b0dbbfeb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 3d 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.

AGENTS.md · 48 lines

How it starts

The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.

pdf-mcp — Agent Navigation

Reading order

  1. SPEC.md — Full spec, architecture, and the Intelligence 2.0 feature set
  2. pdf_mcp/server.py — Entry point, FastMCP app, Starlette HTTP + REST API
  3. pdf_mcp/tools/ — Portmanteau tool modules (extract, manipulate, annotate, forms, convert, validate, rag, intel, agent, meta)
  4. pdf_mcp/services/ — Core engines (Extractor, Manipulator, Converter, Chunker, RagStore, Intel, Llm)
  5. webapp/src/pages/ — React pages (Dashboard, Workbench, Pipeline, Chat, Tools, Skills, Logs)

Key files

File Purpose
run_server.py Dual-transport entry point (stdio / HTTP)
pdf_mcp/server.py FastMCP app, Starlette HTTP app, REST endpoints, recipes, watch-folder, share links, stats
pdf_mcp/config.py Env-based configuration
pdf_mcp/tools/__init__.py Tool module registration
pdf_mcp/tools/agent.py pdf_do agentic chaining tool
pdf_mcp/tools/intel.py pdf_analyze, pdf_redact, pdf_classify, pdf_dedupe, pdf_export
pdf_mcp/tools/meta.py pdf_help, pdf_status, pdf_shutdown
pdf_mcp/services/intel.py Analyzer, Redactor, Classifier, Deduper, BriefBuilder
pdf_mcp/services/llm.py Local LLM discovery + chat completions (Ollama / LM Studio)
pdf_mcp/services/extractor.py PyMuPDF text/image/table/metadata extraction
pdf_mcp/services/manipulator.py pypdf merge/split/rotate/compress/encrypt
pdf_mcp/services/converter.py PDF to/from Markdown/HTML/images
pdf_mcp/services/chunker.py Text chunking strategies (incl. table-aware)
pdf_mcp/services/rag_store.py LanceDB vector store wrapper
webapp/src/App.tsx React router + lazy-loaded pages
webapp/src/pages/Dashboard.tsx KPI dashboard + usage stats panel
webapp/src/pages/Workbench.tsx PDF.js viewer + tool palette + compare mode
webapp/src/pages/Pipeline.tsx Batch jobs + recipes + share links
webapp/src/pages/Chat.tsx LLM chat + PDF search / sources panel
webapp/src/lib/api.ts Fetch layer (REST + RAG search + share + stats)
webapp/src/lib/store.ts Zustand store (backend + LLM provider/model state)

Read the full file on GitHub · 48 lines

Changes

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.

  1. 3d ago First seen · 48 lines · 805 tokens per session scan A b3d5b0dbbfeb

Subscribe to this mod's changes

pdf-mcp AGENTS.md is an instructions file published in the GitHub repository sandraschi/pdf-mcp (1 stars, last pushed 15d ago), licensed MIT. It adds 805 tokens to every session, about $0.0040 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.