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 MuddySheep/vibeguard-local --skill agent-skillgit clone --depth 1 https://github.com/MuddySheep/vibeguard-localWrote 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/muddysheep/vibeguard-local/agent-skill)<a href="https://agentmods.dev/skills/muddysheep/vibeguard-local/agent-skill"><img src="https://agentmods.dev/badge/skills/muddysheep/vibeguard-local/agent-skill/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/muddysheep/vibeguard-local/agent-skill"><img src="https://agentmods.dev/badge/skills/muddysheep/vibeguard-local/agent-skill.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 | $0.00268 | $0.02278 |
| Opus 5.5 | $0.00107 | $0.00911 |
| Sonnet 5.5 | $0.00054 | $0.00456 |
| Haiku 4.5 | $0.00027 | $0.00228 |
Grade A, and why
vibeguard-sql-safety 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 2mo 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 — 225 lines — stays where its author put it; the contents beside it link to each section on GitHub.
vibeguard-sql-safety
You are about to write or execute SQL. Before any SQL touches a real database — production, staging, or even local — you MUST run it through the VibeGuard local SQL safety analyzer. The analyzer is fast (sub-millisecond on typical queries), local-only (no network call), and distinguishes "stop, this is dangerous" from "warning, but probably OK" from "fyi" so you can act with confidence.
When this skill fires
This skill fires when you are about to:
- Execute SQL through any database tool, MCP server, or library
(
pg,psql,prisma migrate,drizzle-kit, etc.). - Propose SQL to the user in a code block they might copy-paste.
- Commit a migration file under
migrations/,db/migrations/,prisma/migrations/, or similar. - Edit a seed / fixture file with INSERT / UPDATE / DELETE statements.
- Construct dynamic SQL through string concatenation, template literals, or any non-parameterized path.
If the statement is parameterized with placeholders ($1, ?, named
binds), still analyze the statement shape — VibeGuard catches issues
at the structure level (missing WHERE, unbounded JOIN, etc.), not at
the literal-value level.
The pre-flight contract
For SQL that lives on disk, run:
vg-local analyze <path-to-sql-file> --format=jsonl
For SQL that is in-memory (the common agent case), pipe it via stdin — no temp file needed:
echo "$SQL" | vg-local analyze --stdin --format=jsonl
Or with a heredoc:
vg-local analyze --stdin --format=jsonl <<'SQL'
<the SQL you are about to run>
SQL
The <stdin> literal appears in the file field of the JSONL output
for stdin reads. --stdin is mutually exclusive with positional file
patterns and with --fix / --fix-dry-run — pick one input source
per call.
The analyzer writes one JSON object per catch to stdout. Each line is a
self-contained record with a stable schema (_schema: "vg-jsonl/1").
If vg-local exits with code 0, there are no block-severity catches.
You may proceed.
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.
- 2mo ago First seen · 225 lines · 268 tokens per session scan A 851f2feb542d
vibeguard-sql-safety is a skill published in the GitHub repository MuddySheep/vibeguard-local (9 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 268 tokens to every session and 2,278 once invoked, about $0.0011 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 2026-08-31.
Other skills, from other repositories
whodb
Work with WhoDB platform hosts and project resources via MCP, including multiple environments, or query databases directly in database-only MCP mode.
schema-designer
Help design database schemas, create tables, and plan data models. Activates when users ask to create tables, design schemas, or model data relationships.
graphjin-eval
Create, extend, run, baseline, and diagnose GraphJin agent evaluations through the graphjin eval CLI.
graphjin-env
Use when setting up a training or evaluation loop against a GraphJin agent environment — running the container, reading /health, driving episodes hosted or step-by-step or with your own agent over MCP, splitting train from eval, exporting trajectories, and deciding whether two rewards can be compared.
slides
Build a Grida slides deck — a .canvas bundle in slides mode whose pages are SVG documents (16:9, one SVG per slide). Use when creating a presentation, pitch deck, slideshow, or talk.
dotcanvas
Author and edit a Grida .canvas board — a .canvas.json manifest plus document files (references, generated images, notes) placed on an infinite canvas. Use when working on a .canvas bundle or arranging visuals/design work spatially. For a linear deck/presentation, use the slides skill instead.