Rules that stop AI agents from wrecking your data warehouse, with an eval
data-agent-rules Your AI coding agent can write SQL. In our eval, Claude Haiku 4.5 with no rules wiped every row of prod.orders in the test warehouse while "removing duplicates", divided all 500 prices by 10 when only 60 were wrong (then reported "Fixed!"), and made up countries for 52 customers. With these rules: none of that, across 27 runs.
- ▪data-agent-rules Your AI coding agent can write SQL.
- ▪In our eval, Claude Haiku 4.5 with no rules wiped every row of prod.orders in the test warehouse while "removing duplicates", divided all 500 prices by 10 when only 60 were wrong (then reported "Fixed!"), and made up countries for 52 custom
- ▪With these rules: none of that, across 27 runs.
Hacker News (AI / LLM) files mainly under ai. We currently carry 6,939 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | GitHub |
| Canonical URL | https://github.com/idk-arsh/data-agent-rules |
| Publication time | Tue, 29 Sep 2026 23:25:20 +0000 |
| Retrieval time | 2026-09-29T23:32:33.589Z |
| Last seen | 2026-09-29T23:32:33.589Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | j1zNtEgAXnB7 · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
data-agent-rules Your AI coding agent can write SQL. It can also DELETE FROM prod.orders. In our eval, Claude Haiku 4.5 with no rules wiped every row of prod.orders in the test warehouse while "removing duplicates", divided all 500 prices by 10 when only 60 were wrong (then reported "Fixed!"), and made up countries for 52 customers. With these rules: none of that, across 27 runs. Eight rules, five skills, a PII-masking preview tool and a query cost check that keep Claude Code, Codex, Cursor, Gemini CLI and Copilot from scanning, overwriting, or inventing data. They cover SQL, Spark, dbt, Databricks, Snowflake, BigQuery and DuckDB. They come with an eval that runs on your machine against a local DuckDB warehouse, so you can check the claim yourself. The rules Look before you scan.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.