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/ovid/paad/claude-mdgit clone --depth 1 https://github.com/Ovid/paadWrote 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/ovid/paad/claude-md)<a href="https://agentmods.dev/instructions/ovid/paad/claude-md"><img src="https://agentmods.dev/badge/instructions/ovid/paad/claude-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 | $0.05565 | $0.05565 |
| Opus 5 | $0.02782 | $0.02782 |
| Sonnet 5 | $0.01113 | $0.01113 |
| Haiku 4.5 | $0.00556 | $0.00556 |
Grade A, and why
paad CLAUDE.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 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.
How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — paad
What this project is
This is a Claude Code plugin marketplace hosted at github.com/Ovid/paad. It distributes the paad plugin, which provides skills for architecture analysis, code quality, and development workflows.
After this file is read, announce "CLAUDE.md loaded."
Also, address me as "Ovid" for further verification that you have read this file.
Project structure
preview/paad/ is where all new work lands; plugins/paad/ is what the marketplace distributes, and it is written only by promotion — see "The two trees" below. Everything else is scaffolding around them. paad/code-reviews/ is committed output from paad's own skills. .claude/skills/ holds project-local skills, which the marketplace does not carry but the npx installer does look in; the only thing keeping them out of an npx install is the internal: true flag documented under "Project-local skills" below. package.json is Pi's package manifest, not a Node one — this is not a JS project; it carries the version and the pi.skills path, and nothing else.
kiro_and_antigravity/ and pi/ are produced from plugins/paad/ by make export — from the shipped tree only, so preview work does not reach them until it is promoted. Everything in them is generated — edit the source or the generator, never the output. README deprecates copying straight out of kiro_and_antigravity/ in favor of npx skills@latest add Ovid/paad, but the route still works and is still documented, so a stale export on main is wrong content shipped to real users.
The two trees
Merging used to ship. npx skills@latest add Ovid/paad clones main and takes what is there — plugin.json's version field does not gate it — so anything merged before it had been driven for real reached users, and keeping it away from them meant keeping it off main. Long-lived branches were the tax, and they made releases hard.
preview/paad/ fixes that by making main safe to merge into:
preview/paad/ all new work lands here; always ahead of, or equal to, plugins/
plugins/paad/ what ships; written only by promotion, never hand-edited
kiro_and_antigravity/, pi/ generated from plugins/ only
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.
- 5d ago First seen · 163 lines · 5,565 tokens per session scan A c96ed3c1cb68
paad CLAUDE.md is an instructions file published in the GitHub repository Ovid/paad (110 stars, last pushed yesterday), licensed MIT. It adds 5,565 tokens to every session, about $0.0278 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 instructions, from other repositories
nopua AGENTS.md
Instructions for wuji-labs/nopua, covering nopua · codex 入仓规则, 一 · 项目定位, 二 · 起手必读, 三 · 工程与研究纪律 and 四 · 对外署名.
code-context-graph AGENTS.md
Instructions for tae2089/code-context-graph, covering code-context-graph, mcp server, agent skills (7), code search rules and documentation.
code-context-graph CLAUDE.md
Instructions for tae2089/code-context-graph, covering code-context-graph, mcp 서버, cli skills (4개), 문서 and 개발 규칙.
wgm AGENTS.md
Instructions for agent-frontier/wgm, covering agents.md, build & run, validate (backpressure), operational notes and codebase patterns.
cdd CLAUDE.md
Instructions for drabaioli/cdd, covering cdd (claude-driven development) — claude code context, key references, critical constraints (quick reference), build & test and module layout.
plan-forge database.instructions.md
Database patterns for .NET — Dapper/EF Core, parameterized queries, migration strategy.