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/designpipe/exfig/claude-mdgit clone --depth 1 https://github.com/DesignPipe/exfigWhat 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 | $0.02202 | $0.02202 |
| Opus 5 | $0.01101 | $0.01101 |
| Sonnet 5 | $0.00440 | $0.00440 |
| Haiku 4.5 | $0.00220 | $0.00220 |
Grade A, and why
exfig 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 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 — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TOON Format Convention
Use TOON (Token-Oriented Object Notation) for all tabular data in this file. TOON reduces token usage by 30-60% by declaring fields once in array headers.
format:
syntax: name[count]{field1,field2,...}:
indent: 2 spaces for rows
delimiter: comma between values
example[2]{id,name,status}:
1,Build command,active
2,Test command,active
When adding lists of items (modules, commands, files, etc.), always use TOON tables instead of markdown tables or lists.
Exception: OpenSpec tasks.md — task items MUST use markdown checklists (- [ ]) for openspec parsing.
Context7 for External Libraries
Always use Context7 MCP to look up documentation for external tools and libraries before implementing:
resolve-library-id- find the library IDget-library-docswithtopic- get relevant docs (usemode: codefor API,mode: infofor concepts)- Paginate with
page: 2, 3...if context insufficient
This applies to: Swift packages, CLI tools (mise, hk, swiftlint, etc.), Figma API, and any third-party dependency.
CLAUDE.md
Agent instructions for ExFig - a CLI tool that exports colors, typography, icons, and images from Figma to iOS, Android, and Flutter projects.
Quick Reference
# Build & Test
./bin/mise run build # Debug build
./bin/mise run build:release # Release build
./bin/mise run test # All tests (prefer over test:filter when 3+ files changed)
./bin/mise run test:filter NAME # Filter by target/class/method
./bin/mise run test:file FILE # Run tests for specific file
# Code Quality (run before commit)
./bin/mise run format # Format all (Swift + Markdown)
./bin/mise run format:swift # Format Swift only
./bin/mise run lint # SwiftLint + actionlint
# Docs & Coverage
./bin/mise run docs # Generate DocC documentation
./bin/mise run docs:preview # Preview docs in browser (localhost:8080/documentation/exfigcli)
./bin/mise run coverage # Run tests with coverage report
# Maintenance
./bin/mise run codegen:pkl # Regenerate Swift types from PKL schemas
./bin/mise run generate:llms # Generate llms.txt + llms-full.txt
./bin/mise run setup # Install required tools
./bin/mise run clean # Clean build artifacts
# Run CLI
.build/debug/exfig --help
.build/debug/exfig colors -i exfig.pkl
.build/debug/exfig icons -i exfig.pkl
.build/debug/exfig batch exfig.pkl # All resources from unified config (positional arg!)
.build/debug/exfig fetch -f FILE_ID -r "Frame" -o ./output
.build/debug/exfig download tokens -o tokens.json # Unified W3C design tokens
.build/debug/exfig mcp # Start MCP server over stdio
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 · 178 lines · 2,202 tokens per session scan A ba30c1fceebf
exfig CLAUDE.md is an instructions file published in the GitHub repository DesignPipe/exfig (10 stars, last pushed 1mo ago), licensed MIT. It adds 2,202 tokens to every session, about $0.0110 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
ccpocket CLAUDE.md
Instructions for K9i-0/ccpocket, covering ccpocket, プロジェクト構成, コマンド, bridge server and flutter app.
mobile CLAUDE.md
Claude Code instructions for lichess-org/mobile, covering claude.md, project overview, development setup, initial setup and install dependencies.
OpenNutriTracker copilot-instructions.md
Copilot instructions for simonoppowa/OpenNutriTracker, covering copilot instructions and commits (quick reminder).
OpenNutriTracker CLAUDE.md
Claude Code instructions for simonoppowa/OpenNutriTracker: Canonical agent instructions for this repository live in AGENTS.md.
zeno-mobile-runner AGENTS.md
Instructions for johnmikel/zeno-mobile-runner, covering notes for coding agents, public metadata policy — read before committing, the rule that catches agents out, scanned surfaces and enable the pre-push hook — once per clone.
DPIP AGENTS.md
AGENTS.md instructions for ExpTechTW/DPIP, covering working in this repository, where things are written down, toolchain, running and commits.