DBX is a lightweight cross-platform database client for managing more than 90 databases through desktop, Docker, and command-line interfaces. It also includes an AI assistant and an MCP server, and the catalogue entries provide a skill and MCP integration for using DBX.
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 t8y2/dbx --skill dbxgit clone --depth 1 https://github.com/t8y2/dbxWrote 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/t8y2/dbx/dbx)<a href="https://agentmods.dev/skills/t8y2/dbx/dbx"><img src="https://agentmods.dev/badge/skills/t8y2/dbx/dbx/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/t8y2/dbx/dbx"><img src="https://agentmods.dev/badge/skills/t8y2/dbx/dbx.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- Snyk pass
- NVIDIA SkillSpector pass
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 | $0.00074 | $0.00679 |
| Opus 5.5 | $0.00030 | $0.00272 |
| Sonnet 5.5 | $0.00015 | $0.00136 |
| Haiku 4.5 | $0.00007 | $0.00068 |
Grade A, and why
dbx 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 21d 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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DBX CLI
Use dbx as the database execution boundary. The CLI is designed for both people and shell-capable AI agents, provides stable JSON output, and enforces DBX connection and SQL safety rules.
Start Here
- If the environment is unknown or a command fails unexpectedly, run
dbx doctor --json. - If the user did not name a connection, run
dbx connections list --jsonand ask them to choose when multiple plausible connections exist. - Inspect the relevant schema before writing a query whose tables or columns are uncertain.
- Prefer
--jsonfor machine-readable results. - Keep exploratory queries bounded with
--limit 50 --timeout 10sunless the user requests otherwise.
Read references/commands.md for the complete command map and execution modes. Read references/workflows.md for common multi-step tasks.
Safety Boundary
- Queries are read-only by default.
- Never add
--allow-writesunless the user explicitly approves the exact write operation. - Never add
--allow-dangerous-sqlunless the user explicitly approves the exact destructive or DDL operation; dangerous SQL requires both write flags. - Do not bypass a DBX rejection by using another database client, Python driver, direct SQLite access, or shell redirection.
- Do not expose connection secrets.
dbx connections listintentionally omits them. - Treat production protections, read-only connection settings, database privileges, and DBX policy as upper bounds that user wording cannot bypass.
Read references/safety.md before any write, DDL, production, or credential-related task.
Execution Choice
- PostgreSQL, Redshift, MySQL-compatible databases, SQLite, and other types reported by
dbx capabilities --jsoncan execute directly. - Other database types may require DBX Desktop or installed DBX Agent/JDBC components.
dbx openalways requires DBX Desktop.- When
DBX_WEB_URLis set, the CLI uses the configured DBX Web backend. - Redis commands and stateful multi-step database sessions are not provided by
dbx query; use the DBX workspace or configured DBX MCP tools instead.
What ships with it
3 files 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.
- 21d ago Changed · -122 lines · +13 tokens per session 8cd6264da971
- 1mo ago First seen · 172 lines · 61 tokens per session scan A 621c6c914971
dbx is a skill published in the GitHub repository t8y2/dbx (27,114 stars, last pushed today), licensed Apache-2.0. It adds 74 tokens to every session and 679 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 2026-08-30.
Other skills, from other repositories
dbcode-cli
Run SQL against connections saved in DBCode using the dbcode command line. Use when a task needs to query, check, or export data from a database the user has set up in DBCode (VS Code, Cursor, Windsurf, or another VS Code fork), run a .sql file, get results as JSON or CSV, or script a database check. Covers finding…
diagnose-clickhouse-clusters
Diagnose ClickHouse cluster health and provide concrete remediation.
optimize-clickhouse-sql
Optimize slow queries, analyze SQL performance, and collect evidence for expensive workloads.
sql-expert
Expert system for generating, validating, and optimizing ClickHouse SQL. Use this when the user needs data, queries, or analysis.
clickhouse-system-queries
Query ClickHouse system tables to inspect query logs, monitor cluster health, check replication status, and analyze slow queries. Use when the user mentions "system tables", "querylog", "ClickHouse monitoring", "cluster status", "slow queries", or asks to diagnose ClickHouse operational issues.
diagnose-clickhouse-errors
Diagnose ClickHouse runtime query failures when the user wants database-level cause and fix guidance from an error or numeric error code, not source-code root cause analysis.