Agents · davila7

read-only-auditor: security audit agent that blocks writes and shell commands

A Claude Code agent for security audits that uses frontmatter hooks to block file writes, edits and Bash calls while it runs.

This text was machine-translated from the publisher's own information and has not been checked by a human editor. The publisher's original text is the reference.

This resource was imported automatically from the sources listed below and has not been reviewed by the AI Gate team yet. The description is the publisher's own, as published. Check the repository before installing.

What it does

The agent performs security audits of a codebase or part of one, such as a directory of payment-processing code. Its frontmatter hooks block the Write, Edit and Bash tools for the length of the audit, so the review is intended to leave files unchanged. The publisher describes this as enforced at the hook level rather than relying on instructions alone. Example checks include unencrypted storage of card numbers, logging of card data, insecure TLS configuration and missing input validation on payment fields.

It suits compliance reviews, pre-merge audits, and situations where the review must not interfere with the code under examination and the audit trail matters. Because it is read-only, findings are reported rather than fixed, so any remediation is a separate step.

The agent is run as a Claude Code agent, invoked with a request that names the code to audit, for example a path in the repository. It is defined in a single markdown file that includes the hook configuration and usage examples.

Limitations and requirements

The hooks block Write, Edit and Bash tool calls for the duration of the audit, so the agent cannot modify files or run shell commands during the review. It requires Claude Code to run.

Safety

Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.

  • Runs code or changes an assistant's behaviour

Sources

  • AI Templates (aitmpl) first seen 13 Nov 2026 · last seen 17 Nov 2026 · its page on the source · figures reported by the source: aitmpl downloads 476
    read-only-auditor

    Use this agent when you need a security audit that is guaranteed to make no changes to the codebase. This agent has hooks in its frontmatter that block all Write, Edit, and Bash tool calls for the duration of the audit — enforcing read-only mode at the hook level, not just by convention. Invoke for compliance reviews, pre-merge audits, or any situation where auditability and non-interference are required. <example> Context: A compliance officer needs a security review of payment processing code without any risk of accidental modification. user: "Audit src/payments/ for PCI-DSS compliance issu

    As shown on www.aitmpl.com when we last read it
Not reviewedAgents
davila7

mcp-protocol-specialist (agent by davila7)

An agent from davila7 described as a specialist in the MCP protocol specification and standards.

Software developmentDesign & media
Updated 17 Nov 2026
Not reviewedAgents
davila7

Unity Game Developer Agent

An agent for Unity game development, covering C# scripting, 3D graphics, mobile optimization and cross-platform deployment.

Software developmentCloud & DevOps
Updated 17 Nov 2026
Not reviewedAgents
davila7

3d-artist: 3D art and asset agent for game development

A game development agent for 3D art and asset creation, covering modeling, texturing, animation, asset optimization and technical art for Unity and Unreal Engine.

Design & mediaSoftware development
Updated 17 Nov 2026