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 rules/mbufkin/cloakbox/cloakbox-guardrails.examplegit clone --depth 1 https://github.com/mbufkin/cloakboxWrote 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/rules/mbufkin/cloakbox/cloakbox-guardrails.example)<a href="https://agentmods.dev/rules/mbufkin/cloakbox/cloakbox-guardrails.example"><img src="https://agentmods.dev/badge/rules/mbufkin/cloakbox/cloakbox-guardrails.example.svg" alt="Measured on agentmods" 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 | $0.00354 | $0.00354 |
| Opus 5 | $0.00177 | $0.00177 |
| Sonnet 5 | $0.00071 | $0.00071 |
| Haiku 4.5 | $0.00035 | $0.00035 |
Grade A, and why
cloakbox-guardrails.example 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 5d 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.
What it actually says
Cloakbox guardrails (sensitive data)
Copy this into .cursor/rules/ in any workspace where an agent can reach this
project. These rules are non-negotiable.
Never touch the vault or the decode side
- NEVER open, read,
ATTACH, query, copy, or print the vault (vault.duckdbor your real warehouse) or any file containing real data. - NEVER read the mapping (
mapping.duckdb), the salt (salt.txt), or anything under~/.cloakbox_vault/(the decode side). - NEVER attempt re-identification. Turning tokens back into people is a manual,
human-only step via
decode/decode.py; an agent must never run it or read its inputs/outputs.
Only the sanitized Cloakbox is allowed
- The ONLY data source you may query is the
cloakboxMCP server (read-only), which points atdata/cloakbox.duckdb. It contains tokens (e.g.SUB_…,TCH_…), never real names. - Do not write to it, and do not open the
.duckdbfiles directly as files.
When asked to analyze data
- Use the
cloakboxMCP tools. Report results using the tokens as-is — never speculate about who a token "really is." - If a task seems to require real identities, STOP and tell the user it must be done manually with the decoder; do not work around this.
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.
- 5d ago First seen · 31 lines · 354 tokens per session scan A eecb5de5766b
cloakbox-guardrails.example is a cursor rule published in the GitHub repository mbufkin/cloakbox (0 stars, last pushed 1mo ago), licensed MIT. It adds 354 tokens to every session, about $0.0018 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 cursor rules, from other repositories
selectools-providers
Rules for working with LLM provider adapters in selectools.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
prefer-assertions-over-defensive-checks
Prefer assertions over defensive checks when data is guaranteed to be valid.