security-audit-master

security-audit-master is an agent for Claude Code from Xsaven/vector-memory-mcp. It costs 23 tokens per session (7,247 once invoked), scanned A, a copy of commit-master, MIT.

A security-review agent for MCP servers and systems that store searchable vector representations of data.

In plain words
What is it for?
Use it to review input validation, sanitize queries and content, enforce safe working-directory boundaries, and check Python data-validation rules.
Why use it?
It helps identify unsafe inputs, path escapes, excessive resource use, and other common security weaknesses before they reach production.

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/xsaven/vector-memory-mcp/security-audit-master
Clone the repo
git clone --depth 1 https://github.com/Xsaven/vector-memory-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 security-audit-master

README.md
[![agentmods](https://agentmods.dev/badge/agents/xsaven/vector-memory-mcp/security-audit-master.svg)](https://agentmods.dev/agents/xsaven/vector-memory-mcp/security-audit-master)
Your own site
<a href="https://agentmods.dev/agents/xsaven/vector-memory-mcp/security-audit-master"><img src="https://agentmods.dev/badge/agents/xsaven/vector-memory-mcp/security-audit-master.svg" alt="Measured on agentmods" height="20"></a>
Per session 23 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 7,247 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 84% copy Near-identical to another mod 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 $0.00023 $0.07247
Opus 5 $0.00012 $0.03624
Sonnet 5 $0.00005 $0.01449
Haiku 4.5 $0.00002 $0.00725

Measured 2d ago against content hash d7cbceea721a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-audit-master 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 2d 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.

Origin

This is a copy

84% identical to commit-master — 589 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.claude/agents/security-audit-master.md · 517 lines

How it starts

The opening of the file, as written. The whole thing — 517 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CONFIDENCE: 0.8 | INDUSTRY_ALIGNMENT: 0.8 | PRIORITY: high

Input validation sanitization

Input validation and sanitization for MCP tool parameters.

  • validate-content: Content length ≤ 10K chars, strip dangerous patterns, validate encoding
  • validate-category: Whitelist: code-solution, bug-fix, architecture, learning, tool-usage, debugging, performance, security, other
  • validate-tags: Max 10 tags, alphanumeric + hyphens only, length 1-50 chars per tag
  • validate-query: Non-empty string, max length, sanitize SQL-like patterns
  • validate-limit: Integer range 1-50, prevent resource exhaustion
  • validate-days: Positive integer, prevent invalid retention policies
  • python-type-validation: Use Pydantic models, dataclass constraints, type hints for runtime validation

Path traversal prevention

Path traversal and directory escape attack prevention for file operations.

  • working-directory-validation: Verify working_dir is absolute path, exists, and is accessible
  • path-sanitization: Resolve absolute paths, normalize separators, validate no traversal sequences (../, .\)
  • boundary-enforcement: Ensure all file operations stay within working_dir boundaries
  • symbolic-link-check: Resolve symlinks and verify final path within allowed directory
  • python-pattern: Use pathlib.Path.resolve(strict=True), os.path.commonpath() for validation

Injection attack prevention

SQL injection, command injection, and code injection prevention patterns.

  • sql-injection: ALWAYS use parameterized queries with ? placeholders, NEVER string concatenation or f-strings in SQL
  • command-injection: Avoid shell=True in subprocess, use argument lists instead of strings, sanitize all external inputs
  • code-injection: Validate all dynamic imports, avoid eval/exec, sanitize template inputs
  • python-sqlite-safety: Use cursor.execute(query, params) with tuples, never format strings into SQL
  • content-hash-safety: Use SHA-256 for deduplication, handle hash collisions gracefully

Read the full file on GitHub · 517 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. 2d ago First seen · 517 lines · 23 tokens per session scan A d7cbceea721a

Subscribe to this mod's changes

security-audit-master is an agent published in the GitHub repository Xsaven/vector-memory-mcp (0 stars, last pushed 6mo ago), licensed MIT. It adds 23 tokens to every session and 7,247 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 84% identical to commit-master, differing in 589 lines, and is treated as a copy.