AiMatch

Agent Kit & Repo Audit

Agent Kit gives developers reusable instructions, hooks, prompts, and checks for coding agents. Repo Audit compares a public repository's agent guidance with its code and configuration and delivers a prioritized, source-backed correction report.

Visit website Compare alternatives Own this tool? Claim it
Agent Kit & Repo Audit screenshot

Who it's for

  • Developers using Claude Code, Codex, Cursor, or GitHub Copilot who want repository-specific instructions and tool setup.
  • Repository maintainers who want a static review of agent instructions against source, manifests, and CI configuration.
  • Teams maintaining guidance across three public repositories that need audits and a cross-repository summary.

Not the right fit if…

  • Teams expecting an audit to execute or test their code: Repo Audit is a static review and never runs repository code.
  • Teams looking for the Kit to install a CI workflow: the Kit installs no CI workflow.
  • Teams expecting the Kit to configure Codex hooks: the page states that Codex hooks are not configured by the Kit.

How it fits your workflow

Buy the Kit to download its shared instructions, tool-specific setup, guard hook, prompts, readiness scan, and tests. Adapt the example commands and paths for your repository, then check installed permissions and hooks.

For Repo Audit, submit a public GitHub, GitLab, or Codeberg URL at checkout. The audit checks instruction files against the repository tree and configuration without running code, then sends a PDF and Markdown report with source-backed findings and prioritized fixes.

The free checker can review one pasted AGENTS.md file for scope, permissions, and structure before a purchase.

Pricing

Free: A free AGENTS.md checker lets users paste one file for findings on scope, permissions, and structure.

Agent Kit

€39 one-time

  • · v1.2 download
  • · Free v1.x updates
  • · Shared instructions and tool-specific setup
  • · Guard hook, workflows, and readiness scan

Repo Audit

€149 one-time

  • · Review of one public repository
  • · Up to 10 instruction files and 1,200 instruction lines
  • · Six report sections with source references
  • · PDF and Markdown report by email within 48 hours of payment

Kit + Audit

€169 one-time

  • · Everything in Agent Kit
  • · One public-repository audit with the same stated scope as Repo Audit
  • · Kit download after payment and audit report by email within 48 hours

Team Pack

€399 one-time

  • · Agent Kit
  • · Audits of three public repositories
  • · One cross-repository summary PDF
  • · Audit reports and summary delivered by email within five working days

Prices checked on Nov 8, 2026 from the vendor's site. They can change; confirm before you buy.

Key features

Shared and tool-specific agent instructions
The Kit includes AGENTS.md and CLAUDE.md alongside configuration and rules for Claude Code, Cursor, and GitHub Copilot. It also includes setup notes for the named tools and supports other agents through AGENTS.md.
Guard hooks and permission guidance
The Kit includes a Claude guard hook and Cursor hook configuration, plus guidance on permission controls and where those controls stop. The page notes that support and behavior vary by tool.
Readiness scan and checks
The Kit includes an agent-readiness scan with 14 automated checks, a 20-point checklist, and regression-test files. The page describes the scan as heuristic and distinguishes it from the full checklist.
Workflow prompts
The Kit contains six prompts covering planning, bug fixes, refactoring, review, test backfill, and handoff.
Static repository audit
Repo Audit compares instruction claims with repository files and configuration, including paths, commands, scripts, packages, and tool rules. It never runs the code.
Evidence-backed audit report
The report includes a summary, stale-instruction checks, conflict and load-order findings, length budgets, a 20-point score, and prioritized fixes with file references and effort estimates.

Works with

  • Claude Code
  • Codex
  • Cursor
  • GitHub Copilot
  • GitHub, GitLab, and Codeberg

Limitations to know

  • The standard audit scope is one public repository, up to 10 instruction files and 1,200 instruction lines. For larger repositories, the provider emails the buyer before starting with the proposed scope.
  • Private repository reviews require contacting the provider before purchase.
  • The audit is static and does not run the repository's code.
  • The Kit installs no CI workflow; required CI checks must be configured by the repository.
  • The Kit does not configure Codex hooks, and Codex sandbox and approvals are configured separately.
  • The Claude guard hook and Cursor guard configuration are limited to configured events and text matching; the page says they do not inspect called scripts, resolve symlinks, or cover MCP.

Getting started

The supplied Claude failure-blocking setting requires version 2.1.295 or later, and the guard requires jq. No other system requirements are stated.

  1. Select the Kit, audit, bundle, or Team Pack and complete checkout.
  2. For an audit, provide the public repository URL from GitHub, GitLab, or Codeberg.
  3. Download the Kit after payment and adapt its example commands and paths to your repository.
  4. Use the included readiness scan and checks; if you ordered an audit, watch for the report by email within the stated delivery window.

Frequently asked questions

Can you review private repositories?

Contact the provider before buying to discuss a private repository.

Which agents are covered?

The page names Claude Code, Codex, Cursor, and GitHub Copilot, and says other agents can use AGENTS.md. Loading and scope vary by tool.

Do you run my code?

No. The audit is a static review of instruction files, source, manifests, and CI configuration; the code is never run.

What if my repository has no AGENTS.md?

The audit drafts an AGENTS.md from the code and repository configuration.

Can I get an invoice for my business?

Yes. Enter business details at checkout, and Stripe emails the invoice. The page says no VAT is charged under § 19 UStG.

How does the Kit differ from free templates?

The Kit combines shared instructions, tool-specific setup, a guard hook, workflows, a read-only scan, and regression tests. Users adapt commands and paths; the Audit checks the actual repository and proposes specific corrections.