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 database-safetygit 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/database-safety)<a href="https://agentmods.dev/skills/catpilotai/catpilot-ai-guardrails/database-safety"><img src="https://agentmods.dev/badge/skills/catpilotai/catpilot-ai-guardrails/database-safety/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/database-safety"><img src="https://agentmods.dev/badge/skills/catpilotai/catpilot-ai-guardrails/database-safety.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 | $0.00075 | $0.04137 |
| Opus 5.5 | $0.00030 | $0.01655 |
| Sonnet 5.5 | $0.00015 | $0.00827 |
| Haiku 4.5 | $0.00007 | $0.00414 |
Grade A, and why
database-safety 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 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.
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 — 435 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Baseline
Applies when: Any SQL execution, ORM write, schema operation, migration command, or query-string construction against a real database, especially prod/production/live/customer environments.
Always:
- Require a
WHEREclause on everyDELETE,UPDATE, orMERGE(and ORM equivalents) unless the user has explicitly confirmed "all rows." - Run the matching
SELECT COUNT(*)with the same predicate and show the count before any mutating statement; the count is what the user approves. - Wrap every mutating statement in an explicit transaction (
BEGIN/COMMIT) opened by the agent; never autocommit against a production-class database. - Clear all three migration gates in order: dry-run on a staging clone, backup taken in the same maintenance window, and a tested reversible migration (or approved restore procedure).
- Parameterize every query through driver-level binding; never build SQL by concatenation, f-strings, or ORM raw-query helpers that bypass binding.
- Identify rows by primary key and redact sensitive columns when debugging; never log full rows or copy production data downstream.
Never:
DELETE/UPDATE/MERGEwithout aWHEREclause, or ORM equivalents likeModel.objects.all().delete()orprisma.model.deleteMany({})with an emptywhere.- A mutating statement run in autocommit mode against a production-class database.
- A migration run straight against production with no dry-run, no backup, no rollback.
- SQL built by string concatenation, f-strings, template literals, or raw-query ORM helpers.
- Copying production rows into development, staging, or any lower-controlled environment, "just for testing."
Open the full component before acting: it has the examples, the remediation steps, and the detection patterns.
Why
Database state is the asset that survives every deploy. Compute can be rebuilt from images in minutes; rows cannot be rebuilt from anywhere except the last backup. Most catastrophic AI-assisted incidents at the database layer follow the same three shapes:
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 Changed · +21 lines c4bf1772ee83
- 2mo ago First seen · 414 lines · 75 tokens per session scan A 61e696e0ba44
database-safety is a skill published in the GitHub repository catpilotai/catpilot-ai-guardrails (2 stars, last pushed 8d ago), licensed MIT. It adds 75 tokens to every session and 4,137 once invoked, about $0.0003 per session on Opus 5.5. 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.
database-safety compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| database-safetycatpilotai/catpilot-ai-guardrails | Skill | 2 | 8d ago | MIT | A | 75 | Claude Code |
| chromaNousResearch/hermes-agent | Skill | 262k | today | MIT | A | 12 | Claude Code, Codex |
| muse_dbasgeirtj/system_prompts_leaks | Skill | 71k | yesterday | CC0-1.0 | A | 0 | Claude Code, Codex |
| prisma-driver-adapter-implementationnitrocloudofficial/nitrostack | Skill | 2.6k | yesterday | Apache-2.0 | A | 71 | Claude Code, Codex |
| prisma-postgres-setupnitrocloudofficial/nitrostack | Skill | 2.6k | yesterday | Apache-2.0 | A | 67 | Claude Code, Codex |
| prisma-upgrade-v7nitrocloudofficial/nitrostack | Skill | 2.6k | yesterday | Apache-2.0 | A | 67 | Claude Code, Codex |
| prisma-clinitrocloudofficial/nitrostack | Skill | 2.6k | yesterday | Apache-2.0 | A | 125 | Claude Code, Codex |