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/aram-devdocs/plumb/agents-mdgit clone --depth 1 https://github.com/aram-devdocs/plumbWhat 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 | $0.01100 | $0.01100 |
| Opus 5 | $0.00550 | $0.00550 |
| Sonnet 5 | $0.00220 | $0.00220 |
| Haiku 4.5 | $0.00110 | $0.00110 |
Grade A, and why
plumb 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 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.
How it starts
The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
This file is the tool-agnostic entry point for every AI coding agent working in this repository — Claude Code, Codex, Cursor, Windsurf, and any future assistant that honors the AGENTS.md convention.
The same content is surfaced to Claude Code via the CLAUDE.md symlink at the repo root.
What Plumb is
Plumb is a deterministic design-system linter for rendered websites. It ships as a single Rust binary with two entry points:
plumb lint <url>— CLI for developers and CI.plumb mcp— Model Context Protocol server for AI coding agents.
The complete product specification lives at docs/local/prd.md. Read the PRD before making non-trivial changes. It is the single source of truth for scope, architecture, determinism invariants, output formats, and release targets.
Repository layout
crates/plumb-core/— rule engine, types, determinism guarantees. No I/O, no async, no wall-clock.crates/plumb-cli/— clap-based binary. Only crate allowed to print to stdout/stderr.crates/plumb-mcp/— rmcp stdio server.crates/plumb-cdp/— Chromium DevTools Protocol driver. Only crate permitted to useunsafe.crates/plumb-config/— config file loading + JSON Schema emission.crates/plumb-format/— output formatters (pretty, JSON, SARIF, MCP-compact).docs/src/— mdBook source for https://plumb.aramhammoudeh.com.docs/adr/— architecture decision records.docs/local/— gitignored scratch space for local-only docs (PRD lives here until extracted)..agents/— tool-agnostic AI library (rules, skills, role specs)..claude/— Claude Code runtime (settings, hooks, agents)..claude/rulesand.claude/skillssymlink into.agents/.
Read order
- This file.
docs/local/prd.md— product requirements document..agents/rules/— project-specific rules (determinism, dependency hierarchy, rule engine patterns, MCP tool patterns, testing, documentation)..agents/skills/— reusable skills (humanizer, code review, etc.).CONTRIBUTING.md— human contributor guide; covers commit conventions and the dev loop.
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.
- yesterday First seen · 65 lines · 1,100 tokens per session scan A 0b91b94cabbc
plumb AGENTS.md is an instructions file published in the GitHub repository aram-devdocs/plumb (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 1,100 tokens to every session, about $0.0055 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
design-md-extractor CLAUDE.md
Instructions for jpoindexter/design-md-extractor, covering claude.md, commands, run a single test file, cli usage (after build) and architecture.
Design-Learn CLAUDE.md
Instructions for GalaxyXieyu/Design-Learn, covering claude.md, todo, quick start, server and vscode extension.
delightful-design-system CLAUDE.md
Instructions for kylesnav/delightful-design-system, covering delightful, source of truth, validate, versioning and ecosystem repos.
delightful-design-system AGENTS.md
Instructions for kylesnav/delightful-design-system, covering delightful, source of truth, validate, editing rules and release notes.
design-md-generator copilot-instructions.md
Instructions for jasonhnd/design-md-generator, covering design md generator, full pipeline (one command flow), 1. extract design tokens, 3. validate and 4. generate report + proof.
design-md-generator AGENTS.md
Instructions for jasonhnd/design-md-generator, covering design md generator, full pipeline (one command flow), 1. extract design tokens, 3. validate and 4. generate report + proof.