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/alohays/paper2pr/agents-mdgit clone --depth 1 https://github.com/alohays/paper2prWhat 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 | $0.09031 | $0.09031 |
| Opus 5 | $0.04516 | $0.04516 |
| Sonnet 5 | $0.01806 | $0.01806 |
| Haiku 4.5 | $0.00903 | $0.00903 |
Grade A, and why
paper2pr 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.
How it starts
The opening of the file, as written. The whole thing — 455 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md -- Paper2PR: presentation framework
Project: Paper2PR Presenter: Yunsung Lee Institution: WoRV, MaumAI Branch: main
Core Principles
- Plan first -- enter plan mode before non-trivial tasks; save plans to
quality_reports/plans/ - Verify after -- compile/render and confirm output at the end of every task
- Single source of truth -- Quarto
.qmdis the only source; there is no other export format. Presentations are delivered from the rendered RevealJS HTML - Design principles -- extreme minimalism, big type, centered content; see
.claude/rules/slide-design-principles.md(mandatory for new decks) - Quality gates -- nothing ships below 80/100; a forbidden term or a level-1 heading is a BLOCKER (score 0), see
.claude/rules/quality-gates.md - [LEARN] tags -- when corrected, save
[LEARN:category] wrong → rightto MEMORY.md - English only -- all content committed to git must be in English (see Language Policy below)
Language Policy
All content pushed to GitHub must be in English. This includes documentation, session logs, commit messages, comments, and slide content.
Exceptions:
.claude/skills/and.claude/agents/may contain Korean examples for bilingual functionality- Speaker notes in QMD files are automatically stripped by the git clean filter (see Speaker Notes below)
- A deck that declares
language.slides: koin its own<deck>.deck.yml. That is the supported way to get Korean slides -- per deck, written on purpose. Do not add a path exemption for it; a directory exemption covers every future deck in that directory, including the ones that should stay English. - An English deck may carry up to
language.korean_allowanceHangul characters (term glosses, Wooclap instructions), counted on the staged qmd after the notes filter. The lecture profile allows 300, the others 0; a deck may set its own number.
Enforcement:
- Pre-commit hook (
scripts/check-korean-pre-commit.sh) blocks Korean text in staged files - Exempt paths:
.claude/skills/,.claude/agents/ - git runs a copy at
.git/hooks/pre-commit, and that copy is the wrapperscripts/pre-commit.sh: the speaker-note gate first, then this one. Editing either script does nothing untilbash scripts/setup-git-filters.shreinstalls it --scripts/test_korean_gate.shchecks the installed hook matches before checking anything else
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.
- 3d ago First seen · 455 lines · 9,031 tokens per session scan A dfc98a7aec5a
paper2pr AGENTS.md is an instructions file published in the GitHub repository alohays/paper2pr (5 stars, last pushed 3d ago), licensed MIT. It adds 9,031 tokens to every session, about $0.0452 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
excel-mcp-server CLAUDE.md
Instructions for TangentDomain/excel-mcp-server, covering excel mcp server, 技术栈, 目录结构, mcp 工具 → python api 映射 and sql 核心原则.
academic-ref-inserter CLAUDE.md
Instructions for fuwenhao12/academic-ref-inserter, covering academic reference inserter (academic-ref-inserter), quick start, full pipeline, or step by step and supported formats.
agentic-resume-builder CLAUDE.md
Claude Code instructions for WtotdeD/agentic-resume-builder, covering agentic-resume-builder development guidelines, what this repo is, active technologies, package manager and validation.
fcp-sheets CLAUDE.md
Instructions for os-tack/fcp-sheets, covering fcp-sheets, project overview, architecture, key patterns and commands.
Doc-Mcp CLAUDE.md
Instructions for mosif16/Doc-Mcp, covering claude.md, project overview, build commands, build (requires rust 1.76+) and run tests.
paperless-ngx-mcp-server CLAUDE.md
Claude Code instructions for cbsmiley/paperless-ngx-mcp-server, covering claude.md, project overview, common commands, architecture and entry points.