pad: Instructions file for Claude Code

CLAUDE.md

pad CLAUDE.md is an instructions file for Claude Code from PerpetualSoftware/pad. It costs 11,153 tokens per session, scanned A, original, Apache-2.0.

Development instructions for Pad, a project-management tool for developers and AI agents. It uses a Go backend, a Svelte web interface, and SQLite, a file-based database.

In plain words
What is it for?
Building and testing Pad, developing its web interface, changing its Go backend or CLI, and working with its agent skill.
Why use it?
They provide the project map and standard commands needed to make changes consistently across the backend, web interface, and command-line tool.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions Claude Code; mentions Codex; mentions OpenCode.

This is PerpetualSoftware/pad's own configuration. It tells Claude Code how to work on pad itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything pad configures →

Reuse

Borrowing it

Nothing to install: this file belongs to PerpetualSoftware/pad. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/PerpetualSoftware/pad/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/PerpetualSoftware/pad

Made for: Claude Code.

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 pad CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/perpetualsoftware/pad/claude-md/github.svg)](https://agentmods.dev/instructions/perpetualsoftware/pad/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/perpetualsoftware/pad/claude-md"><img src="https://agentmods.dev/badge/instructions/perpetualsoftware/pad/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for pad CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/perpetualsoftware/pad/claude-md"><img src="https://agentmods.dev/badge/instructions/perpetualsoftware/pad/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 11,153 This file is loaded in full into every session.
When invoked 11,153 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.11153 $0.11153
Opus 5.5 $0.04461 $0.04461
Sonnet 5.5 $0.02231 $0.02231
Haiku 4.5 $0.01115 $0.01115

Measured yesterday against content hash b99da00ebd6e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

pad 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 yesterday.

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.

CLAUDE.md · 496 lines

How it starts

The opening of the file, as written. The whole thing — 496 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Pad — Development Guide

What This Is

Pad is a project management tool for developers and AI agents. Single Go binary with embedded SvelteKit web UI, SQLite storage, and multi-agent skill support (Claude Code, Cursor, Windsurf, Codex, OpenCode, Copilot, Amazon Q, Junie).

Related repo: The marketing website (getpad.dev) lives at ../pad-web — a separate SvelteKit site deployed to Vercel.

Architecture

  • Backend: Go (cmd/pad/main.go) → REST API (internal/server/) → SQLite (internal/store/)
  • Frontend: SvelteKit 2 + Svelte 5 (web/src/) → static build embedded in Go binary
  • Data model: Workspaces → Collections (typed with JSON schemas) → Items (structured fields + rich content)
  • CLI: Cobra commands in cmd/pad/main.go, HTTP client in internal/cli/
  • Agent skill: Single natural-language /pad skill in skills/pad/SKILL.md

Build & Install

make build      # Build web UI + Go binary (./pad)
make install    # Build, kill server, install to ~/.local/bin/pad, restart
make build-go   # Build Go only (skip web — faster when only backend changes)
make test       # Run Go tests
make web        # Build web UI only
make dev-web    # Run SvelteKit dev server (hot reload on :5173)

After making changes, always run make install to rebuild the binary, install it, and restart the server. The web UI at http://localhost:7777 will reflect the changes.

Quick iteration loop

  • Backend only: make install (skips web rebuild if no frontend changes — edit Makefile to use build-go instead of build in the install target)
  • Frontend only: make web && make install or use make dev-web for hot reload during development
  • Full rebuild: make install

Working in a git worktree

Agent sessions take a git worktree per task rather than sharing the main checkout (a shared checkout means a shared stash stack, branch state, and dirty files across sessions). Three rules keep web tooling working there:

  • Give each worktree its own web/node_modules: run npm ci in its web/. Do not symlink the main checkout's copy. It installs from the npm cache in about 5s warm, costs about 470 MB of disk per live worktree, and runs svelte-kit sync as a side effect. A symlink tests the worktree against whatever the MAIN checkout last installed. web/package-lock.json changes every few days, and the main checkout is not kept current, so that is routinely the wrong dependency set: on day 86 (2026-10-05), worktrees borrowed Tiptap 3.31.3 after main had moved to 3.31.4, plus a materializer bundle built from it, and the corpus tests failed for reasons that had nothing to do with the change. SvelteKit 3 also writes its generated tsconfig inside node_modules with relative paths, so through a symlink a worktree's type check would silently check the main checkout's tree (TASK-3423). (Dave's ruling, 2026-10-06. A pnpm-style shared store is the follow-up if disk becomes the constraint.)
  • A fresh worktree has no web/.svelte-kit (gitignored, generated). Run npx svelte-kit sync in web/ before any vitest/vite command — or npm run check, which syncs first. Without it, vitest fails with Failed to load tsconfig '.svelte-kit/tsconfig.json': Tsconfig not found regardless of how node_modules was set up. (This missing generated dir was historically misdiagnosed as a symlink problem — npm ci "fixed" it only because its prepare script runs svelte-kit sync.)
  • A fresh worktree has no web/build either, and that one breaks the GO gates rather than the web ones. embed.go does //go:embed all:web/build, so make test and make lint fail with pattern all:web/build: no matching files found before a single test runs — go build ./... dies first, two packages report [setup failed], and it reads as a broken tree rather than a missing generated directory. npx vite build in web/ fixes it (make web would too, and is FORBIDDEN here — see the next rule). It also has no web/build-materializer (TASK-2198, gitignored, generated): embed.go does //go:embed web/build-materializer/materializer.js, and a missing file fails go build ./... the same way, with pattern web/build-materializer/materializer.js: no matching files found. Run node scripts/build-materializer.mjs in web/ after npx vite build (npm run build does both). It is built with esbuild, which is a devDependency of its own (vite 8 does not bring it), so a web/node_modules older than the lock file that added it cannot build the bundle. Unlike web/build, a placeholder file here is not enough for make test: internal/materialize's tests run the real bundle.

Read the full file on GitHub · 496 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. yesterday Changed · +44 tokens per session b99da00ebd6e
  2. 3d ago Changed · +19 lines · +208 tokens per session 200531452aa0
  3. 4d ago Changed · +1 lines · +22 tokens per session 53098e5f3749
  4. 7d ago Changed · +14 tokens per session 03f1724a4ac8
  5. 18d ago Changed · +4 lines · +114 tokens per session e3fd81152482
  6. 19d ago Changed · -97 lines · -26,540 tokens per session scan B → A 5e9ccbfe219e
  7. 20d ago Changed · +14 lines · +1,765 tokens per session 9c176e3fa787
  8. 21d ago Changed · +7 lines · +1,126 tokens per session 5f4bdc90eb9a
  9. 22d ago Changed · +107 tokens per session b86a3d6d8f3f
  10. 24d ago First seen · 548 lines · 34,293 tokens per session scan B 834e4b50dd7a

Subscribe to this mod's changes

pad CLAUDE.md is an instructions file published in the GitHub repository PerpetualSoftware/pad (198 stars, last pushed yesterday), licensed Apache-2.0. It adds 11,153 tokens to every session, about $0.0446 per session on Opus 5.5. 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.

Related

Other instructions, from other repositories