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/d-padmanabhan/agent-engineering-handbook/482-snowflakegit clone --depth 1 https://github.com/d-padmanabhan/agent-engineering-handbookWrote 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/d-padmanabhan/agent-engineering-handbook/482-snowflake)<a href="https://agentmods.dev/rules/d-padmanabhan/agent-engineering-handbook/482-snowflake"><img src="https://agentmods.dev/badge/rules/d-padmanabhan/agent-engineering-handbook/482-snowflake.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 | $0.00027 | $0.00755 |
| Opus 5 | $0.00014 | $0.00378 |
| Sonnet 5 | $0.00005 | $0.00151 |
| Haiku 4.5 | $0.00003 | $0.00076 |
Grade A, and why
482-snowflake 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 2d 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 — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Snowflake Engineering Ruleset
Audience: engineers building datasets and pipelines on Snowflake
Goal: secure, cost-aware, high-performance analytics with predictable operations
Warehouses (compute) and cost defaults
- Enable auto-suspend and set an explicit low idle timeout.
- Prefer right-sized warehouses; scale out (multi-cluster) only for concurrency needs.
- Use resource monitors (account/object) to prevent runaway spend.
[!IMPORTANT] Warehouse size is a performance lever and a cost lever. Treat sizing as an explicit decision, not a default.
Security and governance (RBAC-first)
- Prefer roles over users; assign users/service principals to roles.
- Use least privilege:
- separate read roles from write/admin roles
- separate raw vs curated vs serving schemas
- Use platform features for sensitive data:
- masking policies
- row access policies
- secure views
- network policies (where applicable)
[!CAUTION] “Just grant on the database” is usually a governance smell. Scope grants to schemas/tables needed.
Query performance: design for pruning
- Avoid
SELECT *on wide tables in production queries. - Make filters selective and prune-friendly (for example, time ranges).
- Prefer joins on properly typed keys; avoid string casting in join predicates.
- Use clustering keys only when there is evidence of persistent pruning issues.
Anti-patterns to avoid
- Leading-wildcard pattern matching (
ILIKE '%foo') on large datasets. - Unbounded joins across huge tables without predicates.
Data modeling and change management
- Use clear layering (raw/bronze, curated/silver, serving/gold).
- Treat breaking schema changes as migrations:
- dual write / dual read
- versioned objects when needed
- clear rollback plan
Loading data (stages, COPY, Snowpipe)
- Prefer
COPY INTOwith explicit file formats and error handling policies. - For continuous ingestion, use Snowpipe with monitoring/alerting on lag and errors.
- Keep ingestion idempotent:
- stable file naming
- load history checks
- merge/upsert downstream if duplicates possible
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.
- 2d ago First seen · 114 lines · 27 tokens per session scan A 4fef142a212a
482-snowflake is a cursor rule published in the GitHub repository d-padmanabhan/agent-engineering-handbook (16 stars, last pushed 6d ago), licensed MIT. It adds 27 tokens to every session and 755 once invoked, about $0.0001 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 2026-09-03.
Other cursor rules, from other repositories
cdk-infrastructure-data
CDK DynamoDB, SNS, SQS event source, DynamoDB Streams, RDS Proxy, idempotency table.
rds-proxy
RDS Proxy patterns - PostgreSQL connection pooling, Prisma integration, transaction handling, read replicas.
aws-rds-best-practices
AWS RDS PostgreSQL best practices - database configuration, connection management, authentication, backup, and performance optimization standards.
neon-api-projects
Use these rules to manage Neon projects programmatically, including creating, listing, updating, and deleting projects.
neon-api-endpoints
Use these rules to manage compute endpoints associated with branches in a project.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.