pgguard-mcp: Skill for Claude Code

.claude/skills/run-attack-suite/SKILL.md

run-attack-suite is a skill for Claude Code from Azzaraell/pgguard-mcp. It costs 43 tokens per session (356 once invoked), scanned A, original, MIT.

Instructions for running tests that try hostile SQL queries against a SQL policy gate, which is a filter controlling which database statements are allowed.

In plain words
What is it for?
Use them after changing the SQL gate or policy schema to run the attack tests, check named denial rules, verify allowed-query wrapping, and review coverage gaps.
Why use it?
They help verify that dangerous queries are rejected for the right reasons while safe queries still work, and reveal missing attack cases.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is Azzaraell/pgguard-mcp's own configuration. It tells Claude Code how to work on pgguard-mcp 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 pgguard-mcp configures →

Reuse

Borrowing it

Nothing to install: this file belongs to Azzaraell/pgguard-mcp. 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/Azzaraell/pgguard-mcp/main/.claude/skills/run-attack-suite/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/Azzaraell/pgguard-mcp

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 run-attack-suite

README.md
[![agentmods](https://agentmods.dev/badge/skills/azzaraell/pgguard-mcp/run-attack-suite/github.svg)](https://agentmods.dev/skills/azzaraell/pgguard-mcp/run-attack-suite)
Your own site
<a href="https://agentmods.dev/skills/azzaraell/pgguard-mcp/run-attack-suite"><img src="https://agentmods.dev/badge/skills/azzaraell/pgguard-mcp/run-attack-suite/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 run-attack-suite

Your own site · 80×15
<a href="https://agentmods.dev/skills/azzaraell/pgguard-mcp/run-attack-suite"><img src="https://agentmods.dev/badge/skills/azzaraell/pgguard-mcp/run-attack-suite.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 43 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 356 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 $0.00043 $0.00356
Opus 5 $0.00022 $0.00178
Sonnet 5 $0.00009 $0.00071
Haiku 4.5 $0.00004 $0.00036

Measured 12d ago against content hash 720e82fcb7c8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

run-attack-suite 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 12d 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/skills/run-attack-suite/SKILL.md · 29 lines

What it actually says

run-attack-suite

The attack suite in tests/policy.attacks.test.ts is the enforcement claim of this repo made executable. Any change to the gate or the policy schema must keep it green, and new bypass ideas belong in the suite, not in chat.

Steps:

  1. Run npm test -- tests/policy.attacks.test.ts (vitest). All cases must pass: denials with the expected rule id, and the allow-list cases that prove the gate is not a reject-everything filter.
  2. If a case fails, do not relax the test to match the new behavior unless the change was a deliberate policy decision — a failing allow case usually means the column check got stricter, and a failing deny case means a real hole opened.
  3. Check coverage, not just green: compare the suite against the minimum list in the spec (DML/DDL, SELECT INTO, write CTEs, multi-statement, comment smuggling, table/column scope, CTE shadowing, catalog access, denylisted functions, COPY TO PROGRAM, locking clauses, LIMIT clamping). If a class is missing, add it before finishing.
  4. Sanity-check the wrapped output of one allowed query: it must be SELECT * FROM (... ) AS _pgguard_row_cap LIMIT <maxRows> with the original statement preserved byte-for-byte inside.
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. 12d ago First seen · 29 lines · 43 tokens per session scan A 720e82fcb7c8

Subscribe to this mod's changes

run-attack-suite is a skill published in the GitHub repository Azzaraell/pgguard-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 43 tokens to every session and 356 once invoked, about $0.0002 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 2026-08-31.

Related

Other skills, from other repositories

api-testing

Testing patterns for MCP tool/resource handlers using createMockContext and Vitest. Covers mock context options, handler testing, McpError assertions, format testing, Vitest config setup, and test isolation conventions.

cyanheads/obsidian-mcp-server · 46 tokens

add-test

Scaffold a test file for an existing tool, resource, or service. Use when the user asks to add tests, improve coverage, or when a definition exists without a matching test file.

cyanheads/obsidian-mcp-server · 41 tokens

harness-design-fuzzing

Reference vocabulary for designing instrumented harnesses that drive vulnerability discovery — design classes (trigger-driven vs coverage-driven), tiered scope (T1 isolated function / T2 multi-component / T3 full build), systematic input exploration, the two-coverage distinction (fuzzer-feedback vs audit)…

provos/ironcurtain · 147 tokens

field-test

Exercise tools, resources, and prompts against a live HTTP server via MCP JSON-RPC over curl. Starts the server, surfaces the catalog, runs real and adversarial inputs, measures every call (bytes, token estimate, wall-clock) and weighs the catalog, and produces a tight report with concrete findings and numbered…

cyanheads/obsidian-mcp-server · 93 tokens

email-formatting

Markdown formatting conventions for email summary documents — heading depth, list style, line length, emoji policy, and a mandatory provenance footer. Read this when producing a markdown report that summarizes one or more email messages so the output matches the project's house style.

provos/ironcurtain · 53 tokens

json-schema-emails

Canonical shape of the .workflow/emails/emails.json file passed between the fetch and summarize states — required fields (sender, recipient, subject, date, body), types, and field semantics. Read this whenever you write or read emails.json so producer and consumer agree on the shape.

provos/ironcurtain · 62 tokens