foundry CLAUDE.md

foundry CLAUDE.md is an instructions file for coding agents from ai-supervisor-foundry/foundry. It costs 563 tokens per session, scanned A, original, MIT.

Claude Code instructions for ai-supervisor-foundry/foundry, covering foundry — agent instructions, always-apply behavioral rules, cursor rules (.cursor/rules/) and conditional contexts.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/ai-supervisor-foundry/foundry/claude-md
Clone the repo
git clone --depth 1 https://github.com/ai-supervisor-foundry/foundry

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 foundry CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/ai-supervisor-foundry/foundry/claude-md.svg)](https://agentmods.dev/instructions/ai-supervisor-foundry/foundry/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/ai-supervisor-foundry/foundry/claude-md"><img src="https://agentmods.dev/badge/instructions/ai-supervisor-foundry/foundry/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 563 This file is loaded in full into every session.
When invoked 563 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.00563 $0.00563
Opus 5 $0.00282 $0.00282
Sonnet 5 $0.00113 $0.00113
Haiku 4.5 $0.00056 $0.00056

Measured yesterday against content hash 65ec7db38643, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

foundry 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

CLAUDE.md · 41 lines

What it actually says

Foundry — Agent Instructions

Read supervisor-contexts/CONTEXT.md first — it indexes all system documentation. For each *.md or context file, first read the frontmatter (Lines between "---" in the starting of the file) by doing a head -5 before proceeding - if not required, skip file.

Always-Apply Behavioral Rules

  1. Be concise unless asked for elaboration.
  2. Propose = suggest without edits.
  3. After root cause/fix identified, NEVER run commands—verify I approve.
  4. Check MCP availability before asking.
  5. Questions → answer only, no mutating actions.
  6. Mistake → alert, inform, halt. No changes.

Cursor Rules (.cursor/rules/)

  • always: Concise responses, propose-only, 6-line limit, verify approval, questions=answer-only, halt on mistakes
  • supervisor-specs: Operator goals only, no scope expansion, deterministic validation, no anti-goals
  • task-lifecycle: Blocked tasks never autocompleted — set to pending for supervisor
  • cleanup: Ask before cleanup/delete; use ./tmp for *.baks
  • pm2: Logs always --nostream; lifecycle: halt→stop→rebuild→restart→resume
  • secrets: Never print secrets; shell-based length check only
  • mcp: Tool fail → report and halt
  • restart: Halt→stop→rebuild→restart→resume
  • contexts: Read CONTEXT.md first; use supervisor-contexts/ for system, contexts/ for projects

Conditional Contexts

  • System docs: supervisor-contexts/ (read specific section files as needed)
  • Project details: README.md, docs/*.md (not docs/plans unless working on a plan)
  • Sandbox projects: contexts/sandbox/
  • UI Dashboard: UI/ — Express API backend + React frontend; entry point: UI/CLAUDE.md
  • Core source: src/ — see src/CLAUDE.md
  • Tests: tests/ — see tests/CLAUDE.md
  • Scripts: scripts/ — see scripts/CLAUDE.md
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 First seen · 41 lines · 563 tokens per session scan A 65ec7db38643

Subscribe to this mod's changes

foundry CLAUDE.md is an instructions file published in the GitHub repository ai-supervisor-foundry/foundry (11 stars, last pushed 1mo ago), licensed MIT. It adds 563 tokens to every session, about $0.0028 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 2026-09-02.