data-in-prompts

data-in-prompts is a skill for Claude Code from catpilotai/catpilot-ai-guardrails. It costs 57 tokens per session (1,017 once invoked), scanned B, original, MIT.

A data-safety guide for keeping real personal or company records out of prompts, uploads, and tests. It recommends made-up or masked data instead and refuses especially sensitive material such as payment details, identity numbers, health records, passwords, and keys.

In plain words
What is it for?
Use it when testing or demonstrating an AI feature that would otherwise receive customer, employee, invoice, support, or other real records.
Why use it?
Information entered into a chat or app may be copied somewhere it was not approved to go, and removing names does not always stop records from being identified.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions Codex; built for cline.

Part of the catpilot-guardrails plugin — 19 skills shipped together

Good fit Use it when testing or demonstrating an AI feature that would otherwise receive customer, employee, invoice, support, or other real records.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/catpilotai/catpilot-ai-guardrails/data-in-prompts
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.

Any agent
npx skills add catpilotai/catpilot-ai-guardrails --skill data-in-prompts
Clone the repo
git clone --depth 1 https://github.com/catpilotai/catpilot-ai-guardrails

Made for: Claude Code.

Or install catpilot-guardrails, the plugin that ships this one along with the rest of its 19 skills.

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 data-in-prompts

README.md
[![agentmods](https://agentmods.dev/badge/skills/catpilotai/catpilot-ai-guardrails/data-in-prompts/github.svg)](https://agentmods.dev/skills/catpilotai/catpilot-ai-guardrails/data-in-prompts)
Your own site
<a href="https://agentmods.dev/skills/catpilotai/catpilot-ai-guardrails/data-in-prompts"><img src="https://agentmods.dev/badge/skills/catpilotai/catpilot-ai-guardrails/data-in-prompts/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.

agentmods 80×15 button for data-in-prompts

Your own site · 80×15
<a href="https://agentmods.dev/skills/catpilotai/catpilot-ai-guardrails/data-in-prompts"><img src="https://agentmods.dev/badge/skills/catpilotai/catpilot-ai-guardrails/data-in-prompts.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,017 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00057 $0.01017
Opus 5 $0.00028 $0.00508
Sonnet 5 $0.00011 $0.00203
Haiku 4.5 $0.00006 $0.00102

Measured 4d ago against content hash 39e3e4dbda9e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-22, from the pricing page.

Security

Grade B, and why

data-in-prompts 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 4d 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.

Strips warnings and disclaimersmediumAnti-refusal

Omitting safety caveats hides risk from the user and is a common jailbreak preamble.

- Keep it short. Do not lecture. Move to the safe alternative in the same
src/skills/safe-building/data-in-prompts/SKILL.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

When this applies

  • The person is about to paste, upload, or connect real records: a customer export, a spreadsheet of employees, invoices, support tickets, chat logs, a report from the company's sales or HR system.
  • They want "realistic" test data and reach for the real thing.
  • They ask you to "clean up" or "anonymize" a real file so they can use it. Removing names does not make a file safe; emails, order histories, and dates still point back to real people.

What to ask

  • "What is in this file, and whose information is it?" Wait for the answer.
  • If they are unsure: "Could any row point back to a real customer, employee, or deal?"

What to say

  • Name the risk in one sentence: "Once real customer data is in this chat or this app, it has been copied somewhere it was never approved to be, and neither of us can take that back."
  • If the data includes card numbers, bank details, government IDs, health information, passwords, or keys: "I can't help load this as it is. Let's build the same thing with a pretend version first."
  • Keep it short. Do not lecture. Move to the safe alternative in the same message.

Safe alternative

  • Offer to make a sample file with the same columns and made-up rows: invented names, addresses that are obviously fake, emails ending in example.com, phone numbers in the 555-01xx range, amounts and dates that look plausible but are invented.
  • The sample stands in for the real file: point the app, the tests, and the examples at the sample only, never also at the real export, and never paste real rows anywhere, even as an example of the format.
  • Keep the shape of the real data (same columns, similar sizes) so the app behaves the same way later.
  • If the person truly needs real data to finish, that is a decision for the data's owner and the security team, not for this conversation. Tell them who to ask and keep building with the sample in the meantime.
  • Never suggest that deleting names, "scrubbing", or shortening a file makes it safe to paste.

Read the full file on GitHub · 102 lines

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. 4d ago Changed · +3 lines 39e3e4dbda9e
  2. 7d ago First seen · 99 lines · 57 tokens per session scan B a493abe28dd4

Subscribe to this mod's changes

data-in-prompts is a skill published in the GitHub repository catpilotai/catpilot-ai-guardrails (2 stars, last pushed today), licensed MIT. It adds 57 tokens to every session and 1,017 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (strips warnings and disclaimers). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-14.