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 dbt-labs/dbt-agent-skills --skill querying-the-dbt-information-schemagit clone --depth 1 https://github.com/dbt-labs/dbt-agent-skillsWrote 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/dbt-labs/dbt-agent-skills/querying-the-dbt-information-schema)<a href="https://agentmods.dev/skills/dbt-labs/dbt-agent-skills/querying-the-dbt-information-schema"><img src="https://agentmods.dev/badge/skills/dbt-labs/dbt-agent-skills/querying-the-dbt-information-schema/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/dbt-labs/dbt-agent-skills/querying-the-dbt-information-schema"><img src="https://agentmods.dev/badge/skills/dbt-labs/dbt-agent-skills/querying-the-dbt-information-schema.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 | $0.00117 | $0.04438 |
| Opus 5.5 | $0.00047 | $0.01775 |
| Sonnet 5.5 | $0.00023 | $0.00888 |
| Haiku 4.5 | $0.00012 | $0.00444 |
Grade A, and why
querying-the-dbt-information-schema 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 23d 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 — 257 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Querying the dbt Information Schema
The dbt Information Schema is a set of SQL views over your project's metadata: models, sources, tests, columns, DAG edges, configs, and run results. You query it with DuckDB SQL, locally, without touching the warehouse.
Use it to answer "what is in this project" questions with one SQL query, instead of grepping YAML or loading a 70 MB manifest.json.
Contents
| You want to… | Go to |
|---|---|
| Check the project can use this | Prerequisites |
| Answer a one-off question | Route A: dbt show (default) |
Run many queries, script them, or use run_results_latest |
Route B: DuckDB on the parquet files |
| Avoid wrong answers from tricky columns | Column gotchas |
| Understand why a view is empty | What is populated when |
| Copy a working query | Common queries |
| Enforce a rule on every build | Turning a query into a check |
Which route? Use dbt show unless you have a reason not to. It always reads the metadata from the latest dbt command, and it needs nothing installed. Switch to DuckDB when it is installed and you will run many queries, because each dbt show call costs about 0.6 s versus about 0.1 s for DuckDB. The trade-off is that you must refresh the parquet files yourself.
Route A: dbt show |
Route B: DuckDB on parquet | |
|---|---|---|
| Table names | {{ info_schema('models') }} (Jinja, bare view name) |
dbt.models, dbt_rt.run_results (after .read views.sql), or 'dbt.models.parquet' |
| Freshness | Updated by build, run, check by default |
Updated only by a command run with --generate-info-schema |
| Needs | dbt v2 | dbt v2 once, plus the duckdb CLI or a parquet library |
dbt_rt.run_results_latest |
Not available | Available |
| Speed per query | ~0.2–0.7 s | ~0.1 s |
Prerequisites
- dbt v2 only. Run
dbt --versionfirst. The docs mark this feature "Available in v2". On dbt Core 1.x, fall back tomanifest.json. - Project metadata must exist.
dbt build,dbt run, anddbt checkwrite it by default.dbt parseanddbt compilewrite it with--generate-info-schema. If none of these has run,dbt show --infofails withInfoSchemaUnavailable(dbt1656) and names a command to run. - Know what the metadata reflects.
select command, generated_at from {{ info_schema('invocations') }} order by generated_at desclists the dbt commands behind it. Mention the latest one in your answer. - If you get 0 rows from a view that should have data, refresh the metadata with
dbt parse --generate-info-schemaand retry.parsedoes not connect to the warehouse.compile,runandbuilddo, so ask before running them. - Only local runs are here. Runs from dbt platform jobs or another machine are not in the local metadata. For those, use the job's own artifacts.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 23d ago First seen · 257 lines · 117 tokens per session scan A e8c7017c8b00
querying-the-dbt-information-schema is a skill published in the GitHub repository dbt-labs/dbt-agent-skills (746 stars, last pushed 3d ago), licensed Apache-2.0. It adds 117 tokens to every session and 4,438 once invoked, about $0.0005 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 2026-10-02.
Other skills, from other repositories
create-pr
Creates a GitHub PR with a Linear-ticket-prefixed title and a decision-led, narrative description for Prisma 8. Use when the user wants to create a pull request, open a PR, or submit changes for review.
schema-exploration
Lists tables, describes columns and data types, identifies foreign key relationships, and maps entity relationships in a database. Use when the user asks about database schema, table structure, column types, what tables exist, ERD, foreign keys, or how entities relate.
atlas-mssql
SQL Server specifics for Atlas (also Azure SQL and MSSQL): login, dev database and URL scope (mode=schema vs mode=database, the database query parameter), compatibility level and collation, system-generated constraint names, clustered index defaults and datetime2 precision in HCL, GO batches and the delimiter…
ha-data-stores
Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…
nornicdb-cypher-queries
Pick fast, predictable Cypher query shapes in NornicDB — point lookups, batch retrieval, pagination, search, traversal, batched UNWIND/MERGE writes, cleanup, multi-tenant isolation. Use when writing or reviewing Cypher whose latency or throughput matters; maps user intent to the executor's hot-path query templates.
volcengine-rds-postgresql
A tool for operating PostgreSQL databases hosted by Volcano Engine's managed database service. PostgreSQL is a relational database used to store structured application data.