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 skills add catpilotai/catpilot-ai-guardrails --skill keys-and-credentialsgit clone --depth 1 https://github.com/catpilotai/catpilot-ai-guardrailsWrote 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/skills/catpilotai/catpilot-ai-guardrails/keys-and-credentials)<a href="https://agentmods.dev/skills/catpilotai/catpilot-ai-guardrails/keys-and-credentials"><img src="https://agentmods.dev/badge/skills/catpilotai/catpilot-ai-guardrails/keys-and-credentials/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/skills/catpilotai/catpilot-ai-guardrails/keys-and-credentials"><img src="https://agentmods.dev/badge/skills/catpilotai/catpilot-ai-guardrails/keys-and-credentials.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 | $0.00053 | $0.00862 |
| Opus 5 | $0.00026 | $0.00431 |
| Sonnet 5 | $0.00011 | $0.00172 |
| Haiku 4.5 | $0.00005 | $0.00086 |
Grade A, and why
keys-and-credentials 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
When this applies
- The tool or app needs to connect to another system: email, a database, a payment provider, a calendar, a sales or HR system, a file store.
- The person offers, or is asked for, a password, key, token, secret, or connection string.
- Generated code or settings contain a real-looking secret.
- A temporary sign-in code or one-time password comes up.
What to ask
- "Does your company have an approved way to connect to this, or a person who sets up connections?"
- If a secret has appeared in the conversation: "Is this the real value?" If yes, treat it as exposed.
What to say
- One sentence: "Anything pasted into a chat or saved in an app can be copied, so a real key here is a key that is already out."
- On being asked for a password: "Don't give it to me or to the app. Let's use the approved connection instead."
- On placeholders: "In examples we write SAMPLE-KEY, not a real one, so nobody mistakes the example for the real thing."
Safe alternative
- Use a built-in connection or company secret store; no key enters chat. Otherwise pause.
- APP_API_TOKEN names a setting, never a secret. Find names in docs; never show credential files. Build without reading its value; let code read it at runtime. Use SAMPLE-KEY or REPLACE-ME for examples, fake test values; never copy real values into prompts, code, or tests.
- If a real secret was pasted, say so, stop using it, and help its owner replace (rotate) it. Deleting the message does not undo exposure.
- Publishable or public keys are a different class. When unsure which kind a value is, treat it as secret.
Company-specific values
Approved services and connection methods:
{{approved_services}}
Stop and ask a human if
- A real password, key, or token has been pasted or saved anywhere.
- The app needs access to payments, banking, HR, or health systems.
- The person plans to share their own login with the app or with others.
- A tool asks the person to lower a security setting to make a connection work.
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.
- 3d ago Changed · +1 lines 05f34806051a
- 7d ago First seen · 87 lines · 53 tokens per session scan A ded5a0523465
keys-and-credentials is a skill published in the GitHub repository catpilotai/catpilot-ai-guardrails (2 stars, last pushed today), licensed MIT. It adds 53 tokens to every session and 862 once invoked, about $0.0003 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 2026-09-14.
Other skills, from other repositories
dcf-model
Build discounted cash flow valuation workbooks in Excel.
baoyu-article-illustrator
Article illustrations: type × style × palette consistency.
flash-attention
Speed up long-sequence transformer training and inference.
subagent-driven-development
Execute plans via delegatetask subagents (2-stage review).
duckduckgo-search
Free keyless web, news, and image search via ddgs.
simpo
Reference-free preference alignment, simpler than DPO.