Borrowing it
Nothing to install: this file belongs to ai-supervisor-foundry/foundry. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ai-supervisor-foundry/foundry/main/.github/instructions/SANDBOX.instructions.mdgit clone --depth 1 https://github.com/ai-supervisor-foundry/foundryWrote 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/ai-supervisor-foundry/foundry/sandbox)<a href="https://agentmods.dev/instructions/ai-supervisor-foundry/foundry/sandbox"><img src="https://agentmods.dev/badge/instructions/ai-supervisor-foundry/foundry/sandbox/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.
<a href="https://agentmods.dev/instructions/ai-supervisor-foundry/foundry/sandbox"><img src="https://agentmods.dev/badge/instructions/ai-supervisor-foundry/foundry/sandbox.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00351 | $0.00351 |
| Opus 5.5 | $0.00140 | $0.00140 |
| Sonnet 5.5 | $0.00070 | $0.00070 |
| Haiku 4.5 | $0.00035 | $0.00035 |
Grade B, and why
foundry SANDBOX.instructions.md scanned grade B with 1 finding 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 1mo 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.
Unrestricted tool accessmediumExcessive agency
A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.
4. After root cause found or fix identified or suspected, NEVER run any commands, verify if I approve of solutions. What it actually says
Always-Apply Behavioral Rules
- Be concise with your ending responses unless asked for elaboration.
- Propose means suggest without edits.
- Don't make more than 6 line changes at a time, if there are more suggest the next 6 lines you would change. After each 6 lines that you have changed, announce, tell me, let me review and acknowledge and then proceed with the next.
- After root cause found or fix identified or suspected, NEVER run any commands, verify if I approve of solutions.
- Always check if you have MCP available before asking me.
- Everytime I ask a question - answer alone and dont take any other mutating actions / make changes.
- Even if you realize you made a mistake. Alert, inform me and halt, dont make changes.
Cleanup Rules
- Always ask me before cleaning up core logic components.
- Always ask me before cleaning up or deleting anything.
- Use ./tmp for *.baks always.
MCP Rules
- If I ask to use an MCP and a tool fails report and halt. Dont proceed.
Secrets Rules
- NEVER print any secret or credentials.
- Always if you have to check, do a shell based length check.
Project CONTEXT
- More about the current project (Timesheet App) here at PROJECT_CONTEXT.md, PROJECT_CONTEXT.md.
- Sub contexts as mentioned in PROJECT_CONTEXT file, are available at contexts.
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.
- 1mo ago First seen · 40 lines · 351 tokens per session scan B 15db6052b00b
foundry SANDBOX.instructions.md is an instructions file published in the GitHub repository ai-supervisor-foundry/foundry (12 stars, last pushed 3mo ago), licensed MIT. It adds 351 tokens to every session, about $0.0014 per session on Opus 5.5. A static security scan graded it B with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-02.
Other instructions, from other repositories
runtm copilot-instructions.md
Copilot instructions for runtm-ai/runtm, covering runtm tool instructions, critical rules, structure, adding endpoints and app/api/v1/myendpoint.py.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).