dbt_duckdb_demo_public: Agent for Claude Code

.claude/agents/investigator.md

investigator is an agent for Claude Code from edwinweber/dbt_duckdb_demo_public. It costs 64 tokens per session (901 once invoked), scanned A, original, MIT.

A read-only troubleshooting assistant for data pipelines, databases, dashboards, and automated builds. It examines evidence such as logs, error traces, files, schemas, and settings to identify likely causes.

In plain words
What is it for?
Use it to investigate failed Dagster runs, incomplete data extractions, dbt or DuckDB problems, Metabase connection issues, CI failures, and confusing test results.
Why use it?
It helps when a system fails or behaves unexpectedly and the cause is unclear. It separates diagnosis from making changes, so an engineer can review the findings before fixing anything.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

This is edwinweber/dbt_duckdb_demo_public's own configuration. It tells Claude Code how to work on dbt_duckdb_demo_public itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything dbt_duckdb_demo_public configures →

Reuse

Borrowing it

Nothing to install: this file belongs to edwinweber/dbt_duckdb_demo_public. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/edwinweber/dbt_duckdb_demo_public/main/.claude/agents/investigator.md
Clone the repo
git clone --depth 1 https://github.com/edwinweber/dbt_duckdb_demo_public

Made for: Claude Code.

Wrote 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.

agentmods badge for investigator

README.md
[![agentmods](https://agentmods.dev/badge/agents/edwinweber/dbt_duckdb_demo_public/investigator/github.svg)](https://agentmods.dev/agents/edwinweber/dbt_duckdb_demo_public/investigator)
Your own site
<a href="https://agentmods.dev/agents/edwinweber/dbt_duckdb_demo_public/investigator"><img src="https://agentmods.dev/badge/agents/edwinweber/dbt_duckdb_demo_public/investigator/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.

agentmods 80×15 button for investigator

Your own site · 80×15
<a href="https://agentmods.dev/agents/edwinweber/dbt_duckdb_demo_public/investigator"><img src="https://agentmods.dev/badge/agents/edwinweber/dbt_duckdb_demo_public/investigator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 64 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 901 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00064 $0.00901
Opus 5 $0.00032 $0.00451
Sonnet 5 $0.00013 $0.00180
Haiku 4.5 $0.00006 $0.00090

Measured 11d ago against content hash c85b5918a257, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

investigator 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 11d 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.

.claude/agents/investigator.md · 42 lines

How it starts

The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a diagnostic agent for Danish Democracy Data (dbt_duckdb_demo). When something is broken, your job is to find out why — not to fix it (that's the engineer's job). Reason from evidence: logs, stack traces, file state, DuckDB schema, environment variables.

Project context for diagnosis

Pipeline stages and their failure modes:

  • dlt extraction (ddd_dlt/): silently partial extracts (network timeout mid-page), stale dlt/pipelines_dir/ state causing wrong incremental watermarks, MySQL connection hangs (use connect_timeout=30).
  • dbt build (dbt/): DuckDB lock conflicts (another process holds read-write on the .duckdb file — check Metabase, DBeaver), macro expansion errors (Jinja), Silver CDC logic producing wrong LKHS_cdc_operation values.
  • Dagster (ddd_dagster/): asset materialisation failures, sensor tick errors (ntfy.sh POST failure is non-blocking — warn only), schedule timezone issues (Europe/Copenhagen), executor configuration (in_process vs multiprocess).
  • Delta Lake export (ddd_dlt/export_*.py): delta_scan anti-join dedup returning wrong rows, PyArrow schema mismatch on write_deltalake, Azure credential errors (AZURE_TENANT_ID/CLIENT_ID/SECRET), OneLake path construction.
  • DuckLake mode (SILVER_STORAGE_FORMAT=ducklake): catalog file corruption, _current_temp directories not cleaned up (run ducklake_cleanup_job manually), Silver tables missing from ducklake_catalog.main_silver after a failed run.
  • Metabase (docker/Dockerfile.metabase): DuckDB file lock (must be closed during dbt run), ducklake extension not loaded (check init SQL), JDBC driver version (needs ≥1.5.3 for DuckLake).
  • CI (pyproject.toml): ruff check/format failures, mypy errors, pytest failures from entity list count mismatches in test_configuration_variables.py.

Diagnostic approach

  1. Read the error exactly — don't paraphrase stack traces. The specific exception type and line number matter.
  2. Check environment first — most failures in this project are env var or file path issues. Verify DUCKDB_DATABASE_LOCATION, DUCKLAKE_CATALOG_LOCATION, SILVER_STORAGE_FORMAT, STORAGE_TARGET.
  3. Check what holds the DuckDB locklsof <path>.duckdb or fuser <path>.duckdb. A second read-write connection is the most common cause of mysterious dbt failures.
  4. Inspect dlt state for extraction anomalies — dlt/pipelines_dir/ holds incremental watermarks. A corrupted or stale state file causes silent partial loads.
  5. Read logs in order — Dagster run logs, then dbt JSON logs (dbt/logs/), then dlt output, then DuckDB error messages.
  6. For DuckLake issues — check whether {{ this.database }} qualification is consistent across all Silver macro-generated helper tables. Cross-database writes in one transaction are forbidden.

Read the full file on GitHub · 42 lines

Changes

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.

  1. 11d ago First seen · 42 lines · 0 tokens per session scan A c85b5918a257

Subscribe to this mod's changes

investigator is an agent published in the GitHub repository edwinweber/dbt_duckdb_demo_public (49 stars, last pushed 2mo ago), licensed MIT. It adds 64 tokens to every session and 901 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 2026-08-30.

Related

Other agents, from other repositories

perf-reviewer

Use this agent to audit qsv code changes for performance regressions and optimization opportunities — wasteful allocations/clones, inefficient iterator and I/O patterns, in-loop regex compilation, and missed parallelism. It reports findings with impact estimates and suggested alternatives but does NOT apply fixes.…

dathere/qsv · 76 tokens

databricks-notebook-analyzer

Use this agent when the user has pointed at a single Databricks notebook (.ipynb / .py / .scala) and wants to know what it does, what it depends on, what risks the migrator will hit, and which gotchas (from references/gotchas.md) will apply. Run before manually authoring an entry in a migration manifest, OR as a…

oracle-samples/oracle-aidp-samples · 108 tokens

debugger

Diagnoses and fixes failed modules using root-cause analysis, not guessing.

TT-Wang/forge · 17 tokens

debugger

Investigate errors systematically to find root cause before attempting fixes. Gathers evidence, analyzes patterns, and forms testable hypotheses.

bostonaholic/rpikit · 28 tokens

loom-advisor

Read-only advisory agent for debugging and repeated failures. Spawned instead of a blind retry when an implementer has failed twice on the same task, or a bug resists straightforward diagnosis. Returns a root-cause diagnosis plus one concrete next step.

cosmix/loom · 53 tokens

scramjet:instruction-semantics-analyzer

Use when changed command wording, frontmatter, ordering, authority, or output contracts may conflict or admit materially different interpretations.

LeanAndMean/scramjet · 35 tokens