MCP Toolbox for Databases is an open-source Model Context Protocol server that connects AI agents, development environments, and applications to enterprise databases through ready-made or custom tools. It is for exploring schemas, querying data, generating code, and building specialized database tools for production agents. The catalogue includes its skills, instructions, and MCP server integration.
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 googleapis/mcp-toolbox --skill reproduce-buggit clone --depth 1 https://github.com/googleapis/mcp-toolboxWrote 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/googleapis/mcp-toolbox/reproduce-bug)<a href="https://agentmods.dev/skills/googleapis/mcp-toolbox/reproduce-bug"><img src="https://agentmods.dev/badge/skills/googleapis/mcp-toolbox/reproduce-bug/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/googleapis/mcp-toolbox/reproduce-bug"><img src="https://agentmods.dev/badge/skills/googleapis/mcp-toolbox/reproduce-bug.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 | $0.00130 | $0.02711 |
| Opus 5 | $0.00065 | $0.01355 |
| Sonnet 5 | $0.00026 | $0.00542 |
| Haiku 4.5 | $0.00013 | $0.00271 |
Grade A, and why
reproduce-bug scanned grade A with 1 finding 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 today.
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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **Transport and client.** Direct `curl`, an SDK, Gemini CLI, an IDE? The bug How it starts
The opening of the file, as written. The whole thing — 217 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reproduce a Bug Report (mcp-toolbox)
Scope: googleapis/mcp-toolbox
About half of reported bugs are not bugs in Toolbox server code: they are misconfiguration, a client-side problem, or something already fixed in a version the reporter has not upgraded to. So the job is not "make it fail", it is to reach a defensible verdict at the lowest cost. Work cheapest-first and stop the moment the evidence settles it. Building a full reproduction for something a one-line version check would have explained is the main way this goes wrong.
The Evidence Rule
Every factual claim MUST cite evidence: a file:line, a commit SHA, a
version string, or the exact command and its output. Otherwise mark it
[UNVERIFIED] and say what would confirm it.
Cite file:line in your findings, where the claim is about the tree as it
is now. This skill's own pointers name files and symbols instead, since line
numbers go stale between releases while symbols survive refactors. Resolve one
with grep -rn '<symbol>' --include="*.go" and read it before citing a line.
Reproducing a bug proves it exists. Failing to reproduce proves much less, and only counts if the attempt was valid, so word the two differently.
Verdicts
Every run ends in exactly one:
| Verdict | Means |
|---|---|
confirmed |
Reproduced on current main. Real, still open. |
already-fixed |
Reproduces on their version, not on main. Needs a release note or a link to the fix. |
misconfiguration |
Their config or environment. |
client-side |
In an SDK, CLI, or LLM client, not the server. Often belongs in another repo. |
works-as-intended |
Deliberate behaviour, frequently a breaking change they have not read. |
not-reproducible |
The attempt was valid and it did not fail. |
blocked |
Needs infrastructure or credentials you do not have. Say exactly what. |
1. Pin the claim down
Read the issue and all comments before touching code.
gh issue view <n> --repo googleapis/mcp-toolbox --json number,title,body,author,labels,comments,createdAt
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.
- today First seen · 217 lines · 130 tokens per session scan A 639d3073abe0
reproduce-bug is a skill published in the GitHub repository googleapis/mcp-toolbox (16,363 stars, last pushed today), licensed Apache-2.0. It adds 130 tokens to every session and 2,711 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-10.
Other skills, from other repositories
langbot-testing
Test LangBot WebUI and core product flows with an automated browser and backend logs. Use when validating the configured LangBot frontend, pipeline Debug Chat, model provider setup and test buttons, bot and knowledge-base UI flows, or troubleshooting failed LangBot end-to-end tests.
app_verification
Verify a newly created or newly installed Rome app after coding:appcreation or coding:workflowcreation finishes installing it. Use only as a second-pass verifier: receive the app id, source root, expected behavior, and local dashboard/API base URL from the creator, then visit the installed app, exercise its safe…
local-test
Build, run, and test IronClaw locally using Docker containers and Chrome MCP browser automation.
examples-qa
Verify Instructor behavior through the repository's ./examples/ suite in pass, live record, or hermetic replay mode. Use when running selected examples or the corpus, capturing and reusing recorded LLM HTTP responses, diagnosing hub results, or distinguishing real errors, assertion failures, skipped examples, and…
knowledge-engineering-quality-and-delivery-real-api-qualification-and-e2e
A guide to tests that call real AI providers and verify complete user flows across the CLI, web interface, terminals, and ACP hosts.
knowledge-engineering-quality-and-delivery-unit-integration-and-shared-test-harnesses
A testing guide for the CLI and web interfaces, covering unit, integration, end-to-end, and real-process test setup with shared fixtures and cleanup.