document-qa

document-qa is a skill for Claude Code, Codex from jezweb/vite-flare-starter. It costs 34 tokens per session (479 once invoked), scanned A, original, MIT.

A document question-answering guide for answering questions about an uploaded PDF, image, or text file. It emphasizes locating evidence in the document and being clear when the answer is missing or uncertain.

In plain words
What is it for?
Use it to find information, explain passages, answer yes-or-no questions, or summarize an attached document.
Why use it?
It reduces the risk of guessing about a document’s contents and helps tie answers to specific pages or sections.

Skill for Claude CodeCodex

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 skills/jezweb/vite-flare-starter/document-qa
Any agent
npx skills add jezweb/vite-flare-starter --skill document-qa
Clone the repo
git clone --depth 1 https://github.com/jezweb/vite-flare-starter

Made for: Claude Code, Codex.

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 document-qa

README.md
[![agentmods](https://agentmods.dev/badge/skills/jezweb/vite-flare-starter/document-qa.svg)](https://agentmods.dev/skills/jezweb/vite-flare-starter/document-qa)
Your own site
<a href="https://agentmods.dev/skills/jezweb/vite-flare-starter/document-qa"><img src="https://agentmods.dev/badge/skills/jezweb/vite-flare-starter/document-qa.svg" alt="Measured on agentmods" height="20"></a>
Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 479 The whole file, excluding the scripts and references it only reads on demand.
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 5.1 $0.00034 $0.00479
Opus 5 $0.00017 $0.00239
Sonnet 5 $0.00007 $0.00096
Haiku 4.5 $0.00003 $0.00048

Measured 5d ago against content hash 6563b29eb01a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

document-qa 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 5d 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.

skills/document-qa/SKILL.md · 43 lines

What it actually says

Document Q&A

When to use

The user has attached a document (or referenced one in their files) and is asking a specific question about its content.

Steps

  1. Identify the document — confirm what file the user is asking about. If multiple files are attached, ask via ask_questions which one to focus on.

  2. Read the document:

    • For attachments in the current message: the file is already in context (use directly)
    • For files in the user's R2 storage: call fs_read with the path
    • For URLs: call browser_markdown or use convertToMarkdown server-side
  3. Answer the question — based on what's actually in the document:

    • Quote specific passages when answering yes/no questions
    • For "what does it say about X?" — find the relevant section, quote or paraphrase
    • For "summarise" — use the summarise-url skill pattern
    • For "find me X" — search the text and report what's there
  4. Cite specifics — when you can, reference the page, section, or paragraph: "On page 3, the report states..."

  5. Be honest about limits:

    • If the document doesn't contain the answer, say so explicitly
    • If the document is unclear, acknowledge it
    • If it's image-based and OCR is poor, note that

Style

  • Direct quotes in quotation marks; paraphrases without
  • Page/section references when available
  • Don't speculate beyond what's in the document
  • If asked your opinion on the document's claims, separate the question of "what it says" (factual) from "is it correct?" (analytical, may need fact-check)

What not to do

  • Don't make up content that wasn't in the document
  • Don't blend prior knowledge with document content without flagging it ("the document says X, though more recent data suggests Y")
  • Don't claim to have read more than was actually extracted (if the document is huge and only part loaded, say so)
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. 5d ago First seen · 43 lines · 34 tokens per session scan A 6563b29eb01a

Subscribe to this mod's changes

document-qa is a skill published in the GitHub repository jezweb/vite-flare-starter (48 stars, last pushed 10d ago), licensed MIT. It adds 34 tokens to every session and 479 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 2026-08-30.