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 skills/atomicbot-ai/atomic-agent/pdfnpx skills add AtomicBot-ai/atomic-agent --skill pdfgit clone --depth 1 https://github.com/AtomicBot-ai/atomic-agentWhat 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 | $0.00056 | $0.01013 |
| Opus 5 | $0.00028 | $0.00507 |
| Sonnet 5 | $0.00011 | $0.00203 |
| Haiku 4.5 | $0.00006 | $0.00101 |
Grade A, and why
pdf 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 2d 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 — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Process PDF files from the terminal. Reading PDF text is best done with the
built-in os.fs.read_document tool (pure-JS, no install). Reach for the CLIs
below only for structural operations: merge, split, page extraction,
PDF↔image rendering, and OCR.
Tooling:
qpdf— merge / split / linearize / encrypt (pure structural ops).poppler—pdfinfo,pdftotext,pdftoppm,pdfimages(inspect + render).ocrmypdf— add a searchable text layer to scanned PDFs (optional).
Setup health check (run first, every session)
Verify with one solo step:
[{ "tool": "os.shell.run", "args": { "cmd": "pdfinfo", "args": ["-v"] } }]
Outcome map:
exit 0+ version →popplerpresent, proceed.- stderr
command not found→ enter Setup playbook → "tools missing".
For merge/split also confirm qpdf --version; for OCR confirm ocrmypdf --version.
Setup playbook (when prerequisites are missing)
OFFER concrete help and EXECUTE the fix yourself — do not dump docs on the user.
tools missing
Reply (solo reply step):
"The PDF utilities are not installed. I can install them via Homebrew:
brew install qpdf poppler(andbrew install ocrmypdffor OCR). Install them?"
On yes:
[{ "tool": "os.shell.run", "args": { "cmd": "brew", "args": ["install", "qpdf", "poppler"] } }]
On Linux use apt-get install qpdf poppler-utils ocrmypdf. If brew itself is
missing, point the user at https://brew.sh/ and stop.
When to use
- "Merge these PDFs", "split pages 3-7", "extract text from this PDF".
- "Convert PDF to images" / "make a PDF from these PNGs".
- "OCR this scanned PDF so it's searchable".
When NOT to use
- Simple text extraction for reading — use
os.fs.read_document(no install). - Editing PDF content/layout — out of scope; guide the user to a PDF editor.
- Filling AcroForm fields programmatically — not covered on v1.
Common operations
All examples invoke os.shell.run. Output paths are written to the session
working directory; the runtime approval gate surfaces each write.
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.
- 2d ago First seen · 92 lines · 56 tokens per session scan A d1e2f0cf7afd
pdf is a skill published in the GitHub repository AtomicBot-ai/atomic-agent (2,438 stars, last pushed today), licensed MIT. It adds 56 tokens to every session and 1,013 once invoked, about $0.0003 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.
Other skills, from other repositories
foundry-hosted-agent-validation
Step-by-step process for validating a Python Foundry hosted agent sample (under python/samples/04-hosting/foundry-hosted-agents/) end to end — running it locally (native runtime and azd ai agent run) and after deploying it to an Azure AI Foundry project with azd. Use this when asked to validate a hosted agent sample.
pdf-processing
Extract text from PDFs, fill forms, and merge documents.
pdf-processing
Process and extract information from PDF documents. Use this skill when the user asks to read, analyze, or extract data from PDF files.
human-summary
Write a human-readable mathematical progress report (compiled PDF) on a project for the operator / the mathematician who posed the problem. This is NOT elaboration (the internal strategy synthesis) and NOT the dashboard. Render from the fact graph's verified statements/proofs into a clean, self-contained report …
Read and extract text from .pdf files, page by page. Use when the user uploads or references a PDF to read or summarize.
press-clip
Turn a live article URL into a press clip that looks like the real coverage — the publication's own logo, fonts, photos and layout kept intact, the ads and clutter removed, and (for a roundup) just the client's section. Renders to PDF. You inspect each site and tailor the removal; the bundled script carries no…